Diagnostic Accuracy of Salivary aMMP-8 Test in Infertile Women and Blood Finding Analysis.

Introduction: The literature reviewed in this paper suggests that infertile patients present worse periodontal status, which may be causative to conception complications. This study aimed to validate an aMMP-8 point-of-care mouth rinse test in the population of women with unexplained infertility and compare it to age-matched fertile women with and without periodontitis. Furthermore, blood sampled inflammatory parameters were analyzed and compared between the two groups. It was hypothesized that the infertile women would present worse periodontal status and a greater number of positive aMMP-8 tests than fertile women, and they would have increased inflammatory blood parameters. Material and methods: The study included 50 healthy norm-ovulatory females aged 25-45 years with strictly defined unexplained (idiopathic) infertility, and 50 healthy norm-ovulatory women of the same age who had conceived and delivered naturally.
Results: The sensitivity and specificity of the test for detecting periodontitis was 84% and 72% in the group of infertile patients, 88% and 68% in the group of fertile patients and 86% and 70% in the overall patient population. Infertile patients with periodontitis had less advanced periodontitis than the control group although this difference was not statistically significant. Blood inflammatory markers were significantly higher in infertile than in fertile women.
Conclusion: This study has shown that infertile patients had better periodontal status and less advanced periodontitis than fertile women of the same age. Therefore, when interpreting the results of aMMP-8 tests for diagnosis of periodontitis, one should keep in mind the periodontal status of the examined population.

Introduction

Periodontitis, a chronic inflammatory disease of the tooth's supporting tissues, has been previously linked with adverse pregnancy outcomes (), in particular with lower birth weight, pre-term birth and an increased risk of preeclampsia in pregnant women (-). It is, however, important to highlight that the strength of this association is limited, and further research on the matter is needed. The available literature data also link periodontitis to later conception and problems with male and female fertility (, , ). Some authors have reported worse periodontal status in infertile patients and suspected that this is a factor that complicates conception (, -).

One of the hypothesized mechanisms on the association between periodontitis, fertility and adverse pregnancy outcomes is the effect of inflammatory mediators associated with periodontitis at the level of the endometrium and embryo implantation site. Furthermore, the systemic inflammatory response that develops in some women with periodontitis may be causative to recurrent miscarriages and infertility ().

Periodontal infection-induced cytokines and proteolytic enzymes destroy the gingival tissues and cause persistent silent systemic inflammation. This is reflected by the activation of matrix metalloproteinase-8 (MMP-8), the salivary enzyme responsible for the destruction of periodontal collagen fibers and is directly associated with periodontal inflammation (, ). The clinical periodontal diagnostic procedure may present as time-consuming, as it includes comprehensive clinical periodontal charting with measurement of various indices and is complemented with the radiographic examination. Point-of-care (PoC) chair-side tests were developed to rapidly detect elevated aMMP-8 in a sampled mouth rinse or gingival sulcus fluid to discriminate periodontally healthy from periodontally diseased individuals (, ). With a high sensitivity of 90% and specificity of 70-85% (, ), the results of PoC tests are comparable with more expensive and complicated tests such as immunofluorometric assay (IFMNA) and Dento ELISA (Enzyme-linked immunosorbent assay). All these tests selectively identify the activated form of MMP-8 (aMMP-8) using the same aMMP-8 antibody and correlate well with each other. A study by Nwhator et al. () has shown a significant association between poor oral hygiene and/or periodontitis and subnormal sperm count. Furthermore, they have demonstrated that the use of MMP-8 PoC chair-side test overcame deficiencies of CPITN index (Community Periodontal Index of Treatment Needs) with reported 96% sensitivity for poor oral hygiene, 95% sensitivity for chronic periodontitis and 82.6% sensitivity for bleeding on probing. However, these numbers were decreased in patients with better oral hygiene or periodontal status. Recently published studies have also shown that the aMMP-8 test has certain limitations and that its sensitivity and specificity increases with the level of periodontitis (, ).

Subclinical chronic systemic inflammation ("low-grade chronic inflammation") is a condition characterized by elevated levels of inflammatory markers such as C-reactive protein (CRP), tumor necrosis factor α (TNF-α), and various interleukins (IL) (). It is considered one of the etiopathogenetic mechanisms of idiopathic infertility (). Several studies have investigated the role of hematological parameters such as complete blood count (CBC), leukocyte count (L), neutrophils (N), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lypmphocyte ratio (PLR) and mean platelet volume (MPV) in the etiopathogenesis of infertility of unknown cause, and their impact on the results of IVF procedures (-). It seems that parameters such as lymphocytes, PLR and MPV could play a role in predicting the success of IVF procedures. Thus, several studies that included subjects treated for idiopathic infertility presented lymphocytes as positive predictive markers for fertilization and PLR as negative markers for implantation rates (, ). Platelet-to-lymphocyte ratio (PLR) and neutrophil-to-lymphocyte ratio (NLR) are cheap and reliable biomarkers of inflammation that have been used in many different diagnoses (, ).

This study aimed to validate an aMMP-8 point-of-care mouth rinse test in a population of women with idiopathic infertility compared to age-matched fertile women with and without periodontitis. In addition, inflammatory parameters in the sampled blood of infertile and fertile women were compared. It was hypothesized that infertile women would have worse periodontal status and a greater number of MMP-8 tests than fertile women, and that they would have increased inflammatory parameters in the blood.

Material and methods

This cross-sectional study was part of a doctoral dissertation study, approved by the Ethics Committee, School of Dentistry, University of Zagreb, Croatia (approval number: 05-PA-15-11/2017.) and institutional review board of "Sestre milosrdnice" University Hospital, Zagreb. The research was registered at the U.S. National Institutes of Health (clinicaltrials.gov) (trial identifier: NCT03542630). Each patient signed informed consent according to the Declaration of Helsinki. The study included 50 healthy norm-ovulatory females aged 25-45 years with strictly defined unexplained infertility attending the reproduction unit of "Sestre milosrdnice" University Hospital, Zagreb, Croatia, and 50 healthy norm-ovulatory women of the same age who conceived and delivered naturally. All subjects attended the Department of Oral Medicine, School of Dental Medicine, University of Zagreb, for clinical periodontal evaluation and aMMP-8 testing between February 2018 and December 2020. The sample size was determined based on the research of Lorenz et al. (). According to the results of their study which have shown the percentage of positive samples of 8.6% (3/35) in the healthy group and 40% (14/35) in the group with periodontitis, it was calculated that 25 samples per each group of subjects were needed to provide an 81% power. The research plan was delivered in oral and written form, and informed consent was received from each participant before the enrolment. Medical and dental histories were obtained from each participant.

Inclusion criteria for women with unexplained infertility were: 1) age between 25 and 45 years; 2) normal ovulatory function; 3) tubal patency; 4) normal semen analysis of a male partner; 5) AMH (Anti-Mϋllerian hormone) values between 15 and 48pmol/l and 6) attempt to conception for the duration of at least one year. Exclusion criteria were specified according to conditions mentioned above and on evidence of autoimmune, metabolic diseases and diabetes since these diseases may increase MMP-8 values. Inclusion criteria for the control group were: 1) age between 25 and 45 years; 2) having conceived and delivered naturally; 3) without evidence of autoimmune, metabolic diseases and diabetes. None of the participants was prescribed or was taking medications such as contraceptives, steroid hormones, insulin-sensitizing drugs, antibiotics and anti-inflammatory drugs that could affect the periodontal status within six months before the inclusion in the study.

Upon confirmation of enrolment eligibility in the study, all participants were screened for periodontitis during a clinical examination. Periodontal examination was performed by two mutually calibrated periodontists (L.M. and I.P.) Plaque index (PI, calculated as full-mouth plaque score – FMPS), bleeding on probing (BOP, calculated as full-mouth bleeding score – FMBS), gingival recession (REC) and periodontal probing depth (PPD) were determined at six sites per tooth, excluding third molars. All parameters were determined using a standard periodontal probe (PCP UNC15, Hu-Friedy Chicago, IL, USA). The staging of periodontitis was determined based on the latest 2018 classification of periodontal diseases and conditions (). Periodontal Inflammed Surface Area (PISA), as an indicator of the systemic inflammatory burden of periodontitis (, ), was calculated for all patients using a calculator within an Excel spreadsheet, available on the website www.parsprototo.info for research use. The calculation was done using CAL (clinical attachement loss), REC (gingival recession) and BOP values for each tooth, as previously described (). A total PISA was calculated through the sum of PISAs of each individual tooth.

Clinical parameters were compared for all patients with positive and negative aMMP-8 test results to perform test validation. Half of the participants were diagnosed with periodontitis in both groups, while the other half of them were diagnosed as those without periodontitis. A matrix metalloproteinase 8 (MMP-8) point-of-care (PoC) chair-side test (Periosafe, Dentognostics GmbH, Jena, Germany) was done from mouth rinse samples. After they had received specific rinsing instructions as specified by the manufacturer, the mouth rinse from participants was filtered into a test cassette and examined after 5 minutes. One line on the test device indicated that the test was performed successfully, and the result was negative. The result was positive if two lines were observed, indicating an elevated risk for periodontitis. The results were observed as positive irrespective of whether the line was light (low-risk), or dark (high risk) in color ().

Biochemical findings from infertile and fertile patients, i.e. complete blood count, differential blood count, blood glucose, erythrocyte sedimentation rate, C-reactive protein, fibrinogen and liver tests, were compared to determine the inter-group differences. Infertile women had their findings done due to processing procedures for infertility. For blood findings, the fertile women were referred to a private laboratory, free of charge.

Statistical analysis

Data were collected in MS Excel spreadsheets and statistical analysis was done in Medcalc (v11.4) program. Data were summarized with a mean (± standard deviation SD) or median and interquartile range (IQR) depending on the normality of the distribution. The normality of the distribution of the collected data was determined by the Kolmogorov-Smirnov test with or without logarithmic data transformation. Continuous variables of normal distribution were compared between the two groups by a t-test. Continuous abnormal distribution data were compared by the Mann-Whitney test between the two groups. Categorical variables were compared by the Chi-square or Fischer's test and values were summed by number and percentage. Correlations between variables were examined using the Spearman test. The level of significance was set at p ˂ 0.05.

Results

The outcomes of the aMMP-8 test were compared with different clinical parameters and diagnoses. The sensitivity and specificity of the test for the detection of periodontitis in the group of infertile patients were 84%, and 72%, in the group of fertile patients 88% and 68% and 86% and 70% in the total patient population (Table 1).

Table 1

Sensitivity and specificity of MMP8 test in the group of infertile and fertile patients in relation to the presence of periodontal pockets ≥4mm and ≥6mm, bleeding index at the level of the mouth ≥15%, periodontal disease (gingivitis or periodontitis) and periodontitis-

Periodontal finding		Infertile (N=50)	Fertile (N=50)	Total(N=100)
Periodontal pocket ≥4 mm	SensitivitySpecificity	65%100%	61.4%83.3%	63.3%90%
Periodontal pocket ≥6 mm	SensitivitySpecificity	85.5%55.6%	85%56.7%	85.3%56%
FMBS ≥ 15%	SensitivitySpecificity	62.2%100%	65.1%71.4%	53.6%83.3%
Periodontal disease	SensitivitySpecificity	74.3%86.7%	87.1%84.2%	80.3%85.3%
Periodontitis	SensitivitySpecificity	84%72%	88%68%	86%70%

The characteristics of infertile and fertile patients based on periodontal findings are shown in Table 2.

Table 2

Characteristics of infertile and fertile patients based on periodontal findings.

	Infertile (N=50)Mean value ±SD / median (IQR)	Fertile (N=50)Mean value ±SD / median (IQR)	P-value
Age	36.1 ± 3.7	37 ± 2.9	*ns
Number of teeth	27 (26-28)	27 (25-28)	*ns
FMBS	38.9 ± 19.7	37.9 ± 19.2	*ns
FMPS	41.6 ± 21.5	36 ± 19	*ns
PISA	645.0 ± 368,0	700.9 ± 420.3	*ns
Number of pockets ≥4 mm	16 (3-29)	22 (6-38)	*ns
Number of pockets ≥6 mm	0 (0-1)	0 (0-6)	*ns
Positive MMP8	28 (56%)	30 (60%)	*ns
Periodontal diagnosis			*ns
Without periodontitis	25 (50%)	25 (50%)
Periodontitis stage I	7 (14%)	0 (0%)
Periodontitis stage II	10 (20%)	10 (20%)
Periodontitis stage III	8 (16%)	15 (30%)

*ns- non significant

In the infertile patients' group, there were 28 positive tests, while in the group of fertile patients, there were a total of 30 positive tests.

Infertile patients with periodontitis presented with a slightly better periodontal status than fertile patients with periodontitis, i.e. they had less advanced periodontitis than the control group. Although the difference was not statistically significant, they had a smaller PISA score and a smaller number of pockets deeper than 4 and 6 mm.

A positive rapid test for MMP-8 significantly correlated with the amount of full-mouth bleeding score (FMBS), the amount of full-mouth plaque score (FMPS), the area of periodontal inflammation (PISA, periodontal inflamed surface area) and the presence of pockets equal to or deeper than 4 and 6 mm (Spearman's correlation coefficient). The correlation was strongest with FMBS, PISA score, and slightly lower with the presence of pockets ≥4 mm, FMPS, and the presence of pockets ≥6 mm.

The sensitivity and specificity of the test depending on the periodontitis stage were also calculated. The result is shown in Table 3.

Table 3

Sensitivity and specificity of MMP8 test in distinguishing subjects without periodontitis and with different stages of periodontitis.

Diagnosis	Positive aMMP-8 test	Negative aMMP-8 test	Total
Periodontitis stage I	7	0	7
Without periodontitis (N=50)	15	35	50
		95% CI (confidence interval)
Test sensitivity	100.00%	59.04% - 100.00
Test specificity	70.00%	55.39% - 82.14%
Periodontitis stage II	15	5	20
Without periodontitis (N=50)	15	35	50
		95% CI (confidence interval)
Test sensitivity	75.00%	50.90% - 91.34%
Test specificity	70.00%	55.39% - 82.14%
Periodontitis stage III	21	2	23
Without periodontitis (N=50)	15	35	50
		95% CI (confidence interval)
Test sensitivity	91.30%	71.96% - 98.93%
Test specificity	70.00%	55.39% - 82.14%

When assessing patients without periodontitis and moderate [stage II] periodontitis, the test's sensitivity was the lowest, - 75%, compared to 91.30% for the advanced [stage III] periodontitis and 100% for initial [stage I] periodontitis.

The results of blood findings are shown in Table 4.

Table 4

Biochemical and clinical parameters of infertile and fertile patients.

	Infertile (N=50)          Mean value ± SD /median (IQR)	Fertile (N=50)          Mean value ±SD /median (IQR)	P-value
Age	36.1 ± 3.7	37 ± 2.9	ns
Body mass index	21.9 (21-23.5)	22 (20.8-23)	ns
Blood glucose	5.3 ± 0,4	5.3 ± 0,5	ns
Hemoglobin	135.4 ± 7	130.6 ± 6,7	0.0007 (t-test)
Hematocrite	0.409 ± 0	0.397 ± 0	0.0079 (t-test)
Erythrocytes	4.5 ± 0.3	4.4 ± 0.2	0.0285 (t-test)
MCH (mean corpuscular hemoglobin)	29.9 (29.4-30.7)	29.7 (28.4-30.4)	ns
MCHC (mean corpuscular hemoglobin concentration)	333 ± 9.9	329.4 ± 11,5	ns
MCV (mean corpuscular volume)	90.2 ± 4.5	90 ± 3.6	ns
RDW (Red blood cell Distribution Width)	12.7 (12.4-13)	13 (12.4-13.2)	ns
Thrombocytes	251.4 ± 46.2	246.9 ± 45	ns
MPV (mean platelet volume)	8 (7.5-8.5)	8.4 (8-9,1)	0.0155 (Mann-Whitney)
Monocytes	5.7 ± 1.3	6.3 ± 1.5	ns
Neutrophils	58.2 ± 8.3	55.1 ± 6.5	0.0439 (t-test)
Leukocytes	7.1 ± 1.8	6 ± 1.2	0.0003 (t-test)
Limphocytes	31.5 ± 7.2	34.2 ± 6.5	0.0478 (t-test)
Bazophils	1 (1-1)	1 (0.5-1)	ns
Eosinophils	2.8 ± 1.6	2.9 ± 1.5	ns
NLR (neutrophil-to-lymphocyte ratio)	1.815 (1.500 – 2.370)	1.58 (1.370 – 1.800)	0.0121 (Mann-Whitney)
PLR (platelet-to-lymphocyte ratio)	8.68 (7.140 – 9.830)	7.07 (6.110 – 7.700)	0.0088 (Mann-Whitney)
Alanine aminotransferase	23 (18-29)	21 (18-24)	ns
Aspartate aminotransferase	17 (15,8-20)	17 (15-20)	ns
Fibrinogen	2.4 (2.2-2.9)	2.3 (2.2-2.4)	ns
Erythrocyte sedimentation rate	6.4 ± 3.9	6.8 ± 4.2	ns
C reactive protein	1.9 ± 1.3	2.7 ± 3.9	ns

Data analysis of infertile and fertile subjects found a statistically significant difference in the values of hemoglobin, hematocrit, erythrocytes, MPV, neutrophils, leukocytes, lymphocytes, NLR and PLR. Infertile patients had significantly higher values of hemoglobin, hematocrit, erythrocytes, neutrophils, leukocytes, NLR and PLR. Fertile patients had significantly higher lymphocyte counts and mean platelet volume (MPV).

Discussion

Some authors have shown worse periodontal status in infertile patients and suspected that this is a factor that complicates conception (, -, ). On the contrary, others have reported that the periodontal status of patients has not affected the outcome of the IVF procedure (). Early diagnosis of periodontitis in the group of infertile patients would enable treatment on time, leading to the elimination of inflammatory stimuli. The papers published after the adoption of the new classification of periodontitis in 2018 () considered the rapid test for aMMP-8 from oral mouth rinse, as used in present research, a key marker for early diagnosis and monitoring disease activity consistent with the new classification (, ).

Our results have shown that sensitivity and specificity of the test for detection of periodontitis in the group of infertile patients was 84%, and 72%, in the group of fertile patients 88% and 68% and 86% and 70% in the total patient population. Rautava et al. () showed lower sensitivity and specificity of the test in patients diagnosed with Crohn's disease, compared with the control group (Crohn's disease group sensitivity 60%, specificity 75%; control group sensitivity 90%, specificity 80%). They concluded that the diagnostic reliability of the test might be compromised in individuals with additional inflammatory conditions due to similar mechanisms underlying disease development. This study is the only published article that has compared the diagnostic reliability of the test in patients with other inflammatory conditions and / or impaired immune responses ().

The results of Heikkinen et al. () showed slightly lower test sensitivity than ours, 76.5% in more than two sites with deep periodontal pockets while specificity was higher than ours, 96.7%. On the contrary, Deng et al. () have shown much lower sensitivity while specificity was similar to ours, 67.1% and 68.8%. They concluded that when performing the test, the number of teeth in the mouth should be considered because with a smaller number of teeth, the concentration of aMMP-8 decreases and the reliability of the test decreases (18 stud). In the present study, patients with a minimum of 20 teeth were included. A significant correlation of the rapid aMMP-8 test with FMBS, FMPS, PISA and the presence of pockets equal to or deeper than 4 and 6 mm was observed, which is consistent with the literature ().

Regarding the test's sensitivity in relation to the staging of periodontitis, it would be expected that the sensitivity of the test increases with the increased stage of periodontitis, as shown by previous authors (, ), however, some of our results deviate from this assumption. Namely, our results showed a test sensitivity of 100% for the detection of mild (stage I) periodontitis, 75% for the detection of moderate (stage II) periodontitis and 91.30% for the detection of severe/advanced (stage III) periodontitis. The explanation probably lies in the small number of patients with mild periodontitis (seven). The results of our study coincide with the published results of Izadi-Borujeni et al. (), who in a study of 60 patients showed a test sensitivity of 87% and a test specificity of 60% in the detection of chronic periodontitis. The same authors calculated the sensitivity and specificity of the test in relation to the severity of periodontitis, and in the case of generalized moderate periodontitis, the sensitivity was 80% and specificity 60%, while in the case of generalized advanced periodontitis, the sensitivity of the test was 93%, with the same specificity. This is consistent with our results showing a test sensitivity of 91.3% and a specificity of 70% when detecting severe periodontitis.

Our research hypothesis was that in patients with idiopathic infertility, we would have a greater number of positive tests, which will be a consequence of a) worse periodontal condition and / or b) increased inflammatory response in the body. Based on the obtained results, this hypothesis has been rejected because it was shown that patients with idiopathic infertility had milder staging of periodontitis than the control group. Thus we can explain a smaller number of positive tests. This is inconsistent with published studies that have shown poorer periodontal status in patients with idiopathic infertility (, , , ). A possible explanation might be that infertile women included in our study were more aware of their oral health, or that they had different economic or educational background which might also affect their oral health, which was not considered in this study. The length of the study is also a limiting factor and it is possible that their oral status was different in the time of establishing the diagnosis of infertility.

In the present study, we have observed a few significant differences between infertile and fertile women regarding the blood findings. Infertile patients had significantly higher values of hemoglobin, hematocrit, erythrocytes, neutrophils, platelets and leukocytes. A possible explanation might be that infertile women frequently take multivitamin preparations, increasing their number of erythrocytes, hematocrit, and hemoglobin. Rudnicka et al. () have showed elevated leukocyte counts and CRP in patients with PCOS and concluded that the main predictive factors for elevated CRP are BMI (Body Mass Index) and IR (insulin resistance). Our results, which have shown only significantly higher leukocytes in infertile patients, but without a significant difference in CRP, are consistent with the literature () because all of our patients had a normal range of BMI, and significant difference in CRP between the study and control group was not observed. Fertile patients had significantly higher mean platelet volume (MPV) and lymphocyte counts. A study of Cakiroglu et al. () proved that MPV values negatively correlated with clinical pregnancy. This is in contrast with our results which have shown higher MPV values in fertile women. We might speculate that their MPV values had been lower before pregnancy. The results from the literature show that lymphocytes are positive predictive markers for fertilization rate (, ), which is in accordance with present results of significantly increased lymphocytes in fertile patients.

The present results have shown that PLR and NLR were significantly increased in infertile patients. In previous studies, NLR and PLR were found to be biomarkers of different inflammatory obstetric conditions, with conflicting results. As reported in the literature, NLR and PLR are increased in women with endometriosis (-) and increased NLR and estradiol-progesterone ratio can be used to predict the development of ovarian hyperstimulation syndrome in patients undergoing controlled ovarian hyperstimulation during in vitro fertilization cycles (). A systematic review has shown that NLR values could be a useful biomarker for predicting preterm delivery (). On the contrary, Yldrm et al. () have demonstrated lower NLR values in patients diagnosed with premature ovarian failure (POI). NLR appears to be a promising marker for POI, one of the possible unrecognized causes of idiopathic infertility, even in the early stages while there is no clinical manifestation of premature menopause yet, and thus can direct clinicians to timely therapy (). PLR, which was significantly increased in infertile women in our study, according to the literature, seems to be a promising marker for predicting implantation failure ().

Our research has some limitations. When including patients in the control group, we did not have a predetermined deadline when the pregnancy occurred, and it is possible that the periodontal status of our patients during the examination was different from the status before the pregnancy.

Conclusion

Our results have shown that infertile patients had better periodontal status and less advanced periodontitis than fertile women of the same age, although they had significantly increased inflammatory blood parameters. These results do not support the assumption that periodontitis-induced inflammatory response may have a role in the etiology of idiopathic infertility. When interpreting the results of aMMP-8 tests for diagnosis of periodontitis, one should consider the periodontal status of the examined population (sensitivity and specificity of the test increase with the staging of periodontitis).