Abstract

Background. Several studies in the last decades have focused on finding a precise method for the diagnosis of periodontal disease in its early stages. Aim. To evaluate from current scientific literature the most common and precise method for gingival crevicular fluid (GCF) sample collection, biomarker analytical methods, and the variability of biomarker quantification, even when using the same analytical technique. Methodology. An electronic search was conducted on in vivo studies that presented clinical data on techniques used for GCF collection and biomarker analysis. Results. The results showed that 71.1%, 24.7%, and 4.1% of the studies used absorption, microcapillary, and washing techniques, respectively, in their gingival crevicular fluid collection. 73.1% of the researchers analyzed their samples by using enzyme-linked immunosorbent assay (ELISA). 22.6%, 19.5%, and 18.5% of the researchers included interleukin-1 beta (IL-1β), matrix metalloproteinase-8 (MMP-8), and tumor necrosis factor-alpha (TNF-α), respectively, in their studies as biomarkers for periodontal disease. Conclusion. IL-1β can be considered among the most common biomarkers that give precise results and can be used as an indicator of periodontal disease progression. Furthermore, paper strips are the most convenient and accurate method for gingival crevicular fluid collection, while enzyme-linked immunosorbent assay can be considered the most conventional method for the diagnosis of biofluids.

1. Introduction

The aim of this review is to evaluate GCF sampling, analytical methods, and oral biomarkers used for the diagnosis and monitoring of different periodontal diseases. It also attempts to explore the main reasons for differences in biomarker quantification among the investigators, by reviewing studies based on inclusion and exclusion criteria from January 2005 to May 2015.

Periodontal diseases are multifactorial infections influenced by the interaction of different types of bacteria with host cells and tissues leading to the release of many cytokines and chemokines which cause the destruction of the periodontal structures [1]. GCF in subjects with periodontal disease contains inflammatory cells, bacteria, tissue breakdown products, antibodies, and complement system proteins and enzymes, in addition to many inflammatory mediators [2]. GCF can be considered among the most nontraumatic investigational methods used to provide information about periodontal tissue conditions, including the status of the connective tissue and the degree of hard tissue destruction [3]. The severity of periodontal tissue inflammation can be estimated by measuring proinflammatory indicators such as IL-1, -6, and -8 and TNF-α that have an important role in the pathogenesis of periodontal diseases [4].

Periodontal diseases must be described by a criteria that is clear and suitable for any examiner to apply, so that the same diagnosis can be reached by other examiners under same conditions [5]. The early diagnosis of periodontal diseases is of significant importance that demands a quick, sensitive, and precise chair-side analytical test [6, 7].

The methods of diagnosis should provide relevant information to assist the differentiation among a variety of periodontal diseases, the degree of periodontal tissue destruction, and the prognosis of periodontal disease. Nowadays, the basic method used in the diagnosis of periodontal diseases depends mainly on the measurement of the clinical periodontal parameters including plaque score (PS) or plaque index (PI), clinical attachment loss (CAL), probing depth (PD), gingival index (GI), or bleeding on probing (BOP), in addition to radio graphical findings. However, the findings from these clinical parameters provide evidence regarding the previous damage of the periodontium rather than clarifying the future condition of the periodontal tissue. Therefore, it is very important to find a method that can predict upcoming periodontal disease. More reliable techniques for periodontal diagnosis need to be researched [105] such as the MMP-8 chair-side testing that depends on the immunochromatography principle which could be beneficial in supporting the clinical diagnostic parameters during the maintenance phase [6, 77]. Moreover, the results from such a test increase the accuracy of the periodontal disease diagnosis and results of unsuccessful treatments [106]. However, this test is fast, inexpensive, and easily performed and takes just 5 minutes [107].

2. Methodology

2.1. Inclusion and Exclusion Criteria

Inclusion criteria in the current study included (a) clinical studies, (b) studies performed on both systemically healthy and systemically unhealthy subjects [patients with diabetes mellitus (DM), coronary heart disease, or rheumatoid arthritis (RA)] with periodontal disease, (c) studies performed on smokers, anxious, or pregnant subjects, (d) studies published only in the English language, and (e) studies clarifying the methodology for GCF sample collection and biomarkers analytical techniques. Exclusion criteria included studies specifically designed to investigate the biomarkers in peri-implant sulcus fluid (PISF), letters to the editor, historic reviews, and commentaries.

2.2. Search Protocol

PubMed, Google Scholar, and Web of Science databases from 2005 to 2015 were searched using combinations of the following keywords: “periodontal disease”, “periodontitis”, “oral biomarkers”, “biomarkers”, and “gingival crevicular fluids”. Titles and abstracts of studies were identified using the above-described protocol, selected by the authors, and checked for agreement. Full texts of the studies were determined by title and abstract and then independently assessed by the authors (Zeyad Nazar Majeed, Dasan Swaminathan, A. M. Alabsi, Koshy Philip, and Saravanan Pushparajan) with reference to the inclusion and exclusion criteria. Initially 2,765 publications were identified, while only 97 studies which fulfilled the inclusion criteria were included and processed for data extraction as shown in the flow chart (Figure 1).

3. Gingival Crevicular Fluid Collection Methods

3.1. Intracrevicular Washing Technique

The device used to perform this method consists of two injection needles placed one inside the other. This method for GCF collection was well described by Salonen and Paunio [108] and the main details of the studies that utilized washing technique in their GCF collection were briefly illustrated in Table 1.

3.2. Microcapillary Technique

A calibrated volumetric or noncalibrated microcapillary pipette with known volume is used to collect GCF. Pradeep et al. [23, 24] gave a good description in their studies on how to perform this collection technique. In this systematic review, 24 studies used this technique (Table 2).

3.3. Absorption Technique

Table 3 summarized the most important findings of the studies that used the absorption technique. Generally, this technique is divided into extracrevicular (Figure 2) and intracrevicular (Figure 3). The first one is performed by placing paper strips over the gingival crevice to reduce trauma. The second method is the intracrevicular technique which is the most commonly used. It may be subdivided into superficial and deep, depending on the depth of strip insertions into gingival sulcus or periodontal pocket [109].

4. Results

A meta-analysis of the results was not carried out in this systematic review because the heterogeneity of the reviewed studies involved the following aspects: GCF sampling, analytical methods, and oral biomarkers used in the diagnosis and monitoring of periodontal disease.

4.1. Sampling Methods

With regard to the techniques used for GCF collection, 97 studies were included and illustrated in our review (Tables 1, 2, and 3). 69 studies used paper strips for GCF collection (Table 3), 24 studies used microcapillary pipettes (Table 2), and only 4 studies utilized the gingival washing method to collect their samples (Table 1).

4.2. Analytical Methods

As shown in Tables 1, 2, and 3 many methods were employed to analyze GCF samples in order to get the desired results. The ELISA analytical method was clearly the preferred method in the majority of the studies. Seventy-one out of the 97 studies that we reviewed used the ELISA technique to analyze the GCF samples.

4.3. Biomarkers

Of the 97 studies, IL-1β (22 studies), MMP-8 (19 studies), and TNF-α (18 studies) were reviewed. These results indicated that these biomarkers concentrations could be used to compare the different stages of periodontal disease and/or assess the effectiveness of periodontal therapy.

Results on sampling and analytical methods and biomarkers are summarized in Figure 4.

4.4. IL-1β

A comparison of GCF IL-1β levels between periodontally diseased patients and healthy subjects was conducted in order to explain the suitability of using IL-1β as a biomarker for periodontal disease progression. In this comparison, only 10 studies among the 22 studies were included (Table 4), due to the exclusion of the studies that did not show enough clinical data and the studies that were conducted on periodontally diseased patients with systemic diseases (e.g., diabetes mellitus) or any other conditions that might affect the results (e.g., pregnant women and smokers). IL-1β concentrations in all studies showed moderate to high significant differences between the healthy subjects and patients with CP or GAgP.

Of the 10 studies that were included to determine the difference in IL-1β levels, only 3 studies [41, 51, 103] showed differences in IL-1β levels between healthy subjects and patients with gingivitis. However this difference was nonsignificant.

4.5. MMP-8

For MMP-8 levels differences between healthy and diseased subjects, only 4 studies fulfilled the inclusion criteria, the same as for IL-1β. All the studies (Table 5) showed differences between healthy subjects and CP patients. Although these differences were highly significant in the studies done by Konopka et al. [67], Rai et al. [88], and Teles et al. [98], it was not considered significant by Yakob et al. [11].

4.6. TNF-α

For TNF-α levels differences between healthy and diseased subjects, only 8 studies were included which fulfilled the inclusion criteria (Table 6). Reis et al. [91] showed that TNF-α levels were significantly higher in diseased sites compared to nondiseased sites. Gokul [14] showed a highly significant difference between healthy subjects and patients with gingivitis and CP. Kurtiş et al. [70] showed highly significant difference between healthy subjects and patients with CP and GAgP. However, the other studies showed only slight difference in TNF-α level between healthy and diseased subjects.

Despite the similarity in results with the majority of studies reviewed in this systematic review, there were still different effects of biomarkers on periodontal disease. The findings of the studies also determined how those biomarkers could be utilized in the diagnosis or monitoring periodontal disease status, as shown in Tables 4, 5, and 6. The present study showed that there was variability in the findings between investigators even when using the same analytical techniques. In order to clarify the differences in outcome, a comparison was made among the studies that used the same analytical methods (Tables 7 and 8).

As shown in Table 7, only five studies were included that used the same analytical method (standard ELISA technique) for quantitative determination of IL-1β concentrations. It was quite noticeable that there were differences in the values of the mean IL-1β concentrations between the studies within the same study group (H, G, CP, and GAgP). For example, the mean IL-1β concentration among healthy subjects in the five studies showed differences in mean values: 49.81, 195.77, 36.44, 15.5, and 17.8732.

Only two studies were involved for comparing MMP-8 mean values (Table 8). Both showed relatively close results in which there was a slight difference in the mean values of MMP-8 in each study group (H, CP).

5. Discussion

This systematic review was designed to discover the most common and accurate GCF collection and biomarker analytical methods and to determine the reliable biomarkers that could be used to detect periodontal disease.

5.1. GCF Sampling Methods

This paper discussed the three main methods of GCF collection: absorption, microcapillary pipetting, and the washing method. There were variations in GCF collection techniques in several of the clinical studies that were reviewed.

5.1.1. Absorption Method

The differences could be summarized as follows:(i)The majority of the studies used paper strips which were considered to be more efficient in GCF collection because they could be inserted easily into gingival sulcus or periodontal pockets, as well as for their ability to absorb fluids. However, few studies used paper points (size 30) to collect GCF samples although it was shown that paper points and paper strips had different absorption rates. A study done by Guentsch et al. [110] indicated that cytokine levels were higher when paper strips were used. Paper points are more commonly used for subgingival plaque collection in microbiological analysis.(ii)The time in which the paper strips or paper points were left in the sulcus varied between 30 seconds [68, 76, 111] and 1 minute [112]. The period most frequently used was 30 seconds to decrease the risk of blood or saliva contamination.(iii)There were variations in the sites from which the GCF samples were collected. Many studies collected GCF samples from diseased sites only [72] in patients with periodontal disease, while other researchers collected samples from both healthy and diseased sites [111].

It was thus important to note that the majority of the studies showed that biomarker levels positively correlated with the periodontal parameters (GI, PD, and CAL). At the same time, it was clarified that healthy sites in individuals with periodontal disease showed increased concentrations of biomarkers in comparison to healthy sites in subjects without periodontal disease. This could be because biomarkers were affected by the bacterial composition of the neighboring subgingival plaque [98] and the fact that the development of periodontitis was site-specific [53].

5.1.2. Microcapillary Pipetting

The time needed to collect GCF samples was related to the desired amount of GCF required and also to the condition of the sample sites (diseased or healthy). The majority of clinical studies collected GCF samples by keeping the microcapillary pipettes at the entry of the pocket for 10 minutes [22, 85]. From our experience, this duration was sufficient if we collected the GCF samples from diseased sites. However, collection of the samples from healthy subjects or healthy sites in patients with periodontitis required 30–50 minutes. This difference in collection time between healthy and diseased sites was due to the flow of GCF which was positively related to the severity of periodontal disease [113115]. The lengthy duration needed to collect GCF when using microcapillary pipettes was considered one of the limitations of this technique, which could increase the possibility of saliva and blood contamination. It would also require more effort from the clinician and could be time-consuming for the subjects.

5.1.3. Washing Method

The results of this review showed that the washing technique of GCF collection were not common due to the technique sensitive difficulties. Also, there was a high rate of blood contamination due to the increased possibility of gingival irritation.

5.2. GCF Analytical Method

In the absence of convincing evidence and the deficiency of data from well-designed studies that focused on the techniques used in the analysis of GCF and after determining the advantages and disadvantages of each technique, it was still difficult to declare that a specific technique was better than others. This was especially so if we considered the following aspects: (a) accuracy and efficiency in biomarker detection and quantification, (b) feasibility, (c) cost, and (d) time. It was clearly shown that the majority of researchers used the ELISA technique in their clinical studies, probably due to its simplicity. It is quite important to clarify that the use of ELISA is not the most accurate technique. For example, Leppilahti et al. [73] showed that the IFMA technique is more accurate than using ELISA.

5.3. Oral Biomarkers for Periodontal Disease Analyzed from GCF Sampling

To date, an accurate diagnosis depends mainly on clinical periodontal examination, radiographic examination, and laboratory tests for microbial analyses [116] that permit a precise evaluation and analysis of bone and attachment loss levels. These findings could be supplemented by GCF analyses where, as many studies have suggested, GCF is a source of bimolecular sampling to investigate the condition of periodontal tissues [112, 117]. GCF ingredients are composed of many components that have been described as markers for periodontal disease development. These comprise host-derived enzymes, host-response modifiers, and tissue breakdown products [64].

It is known that biomarkers are objective and measurable characteristics of biological processes [118] and they can support clinical evaluation, that is, if we fully understand the normal physiology of the biological processes of periodontal disease diagnosis and progression [119]. There are many biomarkers that can be derived from different biofluids such as blood, serum, saliva, and GCF and from different sources such as microbial dental plaque biofilm, connective tissue breakdown products, inflammatory mediators, and host derivatives. For example, MMPs that exist in GCF, saliva, mouth-rinses, and peri-implant sulcular fluid (PISF) can be used to discover a novel chair-side and point-of-care analytical test, which is a nontraumatic method for the diagnosis of periodontal diseases [120, 121].

In this review we focused on GCF biomarkers because of their close proximity to periodontal tissue which minimizes the possibility of reflecting a response on other inflammatory processes in the body.

Several studies have suggested that IL-1β levels can be used as a good biomarker to differentiate between healthy and chronic periodontitis sites [12, 41, 51, 53, 98]. They can also be used to discriminate healthy subjects from patients with AgP [41, 51, 80, 99]. Owing to the slight differences in IL-1β levels between healthy and gingivitis sites [41, 51, 103], it is difficult to use them as indicators or predictors for disease initiation from healthy status to gingivitis.

Much greater levels of MMP-8 in GCF have been observed in periodontitis patients than in healthy subjects [6, 67, 88, 98, 122, 123]. This variation in MMP-8 levels can serve as an indicator for periodontal disease development. Furthermore, Leppilahti et al. [72] found that the levels of MMP-8 in GCF at baseline can predict the behavior of MMP-8 levels during the phase of maintenance.

Yakob et al. [11] however found that there were no statistical differences between healthy and diseased groups, attributing these findings to the differences in the methods used for GCF collection.

The results of this review, as also indicated in many studies, showed minimal increase in TNF-α levels from healthy to periodontally diseased sites [59, 80, 99, 103]. In other studies there was substantial elevation in TNF-α concentration from healthy to diseased sites [14, 70, 91]. Thus, TNF-α concentrations may also be used as a predictor of disease progression.

This systematic review aimed to explore the most reasonable factors that lead to variability in the findings among different studies even when using the same analytical techniques. In order to achieve this, a comparison between the mean values of two biomarkers (IL-1β and MMP-8) was conducted (Tables 7 and 8). The studies that comprised sufficient data such as sample numbers, clear analytical techniques, number or amount of GCF samples, and accurate assessment of the clinical diagnosis through the use of clinical diagnostic parameters (PD, CAL, PI, BOP, and GI) were included in this comparison. Studies that included smokers and diabetic subjects were excluded to minimize the effects on the results. For instance, clinical studies in different populations showed that smoking increased the risk of periodontitis and also that smokers had higher progression and severity of periodontal disease [124]. Tymkiw et al. [102] found that smoking inhibited the expression of many biomarkers including proinflammatory chemokines, regulators of T-cells, and natural killer cells. This inhibition resulted in a decrease in the recruitment of many proinflammatory cytokines and cells to the periodontally inflamed sites, which caused unsuccessful protection against bacterial invasion.

Furthermore, the mechanisms that explain the association between diabetes and periodontitis are not fully understood but encompass aspects of immune function, inflammation, cytokine biology, and neutrophil activity [125]. Types 1 and 2 diabetes have been related to elevated levels of inflammatory mediators [126], such as IL-1β [127] and TNF- α [128].

Table 7 shows a wide variety in the mean values of IL-1β concentrations in the studies. However, the majority of investigators used similar parameters such as size of study population which may affect mean and standard deviation, GCF collection methods (mainly paper strips), analytical techniques (standard ELISA), and clinical diagnostic parameters to categorize the study sample (mainly PD, CAL, PI, and BOP). However, we have noticed that companies manufacturing ELISA kits apply different protocols for measurements of biomarkers and the kit reagents may vary in their detection ability. Another contributing factor to this variability may be the difference in the amount of the collected GCF fluids, ranging from 1–4 paper strips collected from each subject. Such differences in GCF volumes may also cause wide variation in detection rates of biomarkers. This can be supported by the results in Table 8 which showed slight differences between the biomarker mean values, as both studies used the same ELISA kit protocol (R & D System) and an equal number of GCF samples (1 paper strip).

6. Limitations

Due to the heterogeneity of strategies used in the reviewed studies such as sampling, analytical methods, and biomarkers used, a meta-analysis of the results was not possible.

7. Conclusion

In the case of GCF collection methods, paper strips are the easiest and a more precise method. For GCF sample analysis, it is difficult to determine the most accurate method of analysis, but this review has noted that the majority of researchers depended on ELISA technique.

It can be concluded that it is better to use more than one biomarker in determining the inflammatory activity of periodontal disease. IL-1β and MMP-8 can be considered the most preferred cytokines for determining inflammatory activity in the periodontium.

The collected GCF volume and different ELISA kit manufacturing companies are the major causative factors for variation among the investigators.

In general, it is still early to depend on oral biomarkers alone in the diagnosis of periodontal disease, especially in the absence of universal methods for the collection and analysis of these biomarkers. However, it can be utilized to support the clinical parameters which are the most reliable diagnostic methods and also for monitoring periodontal disease progression.

Recommendations for Future Studies

The aim for investigating oral biomarkers is to discover the possibility of using them in the prediction, diagnosis, and monitoring of periodontal diseases or at least to be used as adjunctive to traditional periodontal examination and diagnosis. We believe that, in order to achieve this, researchers should take into consideration the following recommendations in their future studies:(1)The measurement of GCF biomarkers levels should be done by using different collection and analytical methods in order to determine the most accurate technique that can be standardized universally.(2)Comparison of GCF biomarkers levels in different ethnic groups consisting of large sample size should be considered in order to explore the effects of genetic differences on biomarkers levels and also to enable proper statistical analysis.(3)The choice of biomarkers in GCF derived studies is important. Biomarkers that are known to influence the prediction and progression of periodontal diseases should be investigated for statistical correlation to each other. Examples of biomarkers which can contribute to this are IL-1β and MMP-8 whereby Salminen et al. [129] used three salivary biomarkers together to diagnose periodontal disease.

Competing Interests

The authors have declared that there is no conflict of interests.

Authors’ Contributions

All authors contributed equally to this work.

Acknowledgments

This work was supported by the University of Malaya’s High Impact Research with Malaysian Ministry of Higher Education (HIR-MOHE) Research Grant UM.C/625/1/HIR/MOHE/SC/08 (account no. F000008-21001) under the Principal Investigator Koshy Philip and the University of Malaya, Kuala Lumpur, Malaysia, under Grant no. PG226-2014B.