Abstract
Background:
Myeloperoxidase (MPO), the most abundant protein in neutrophils, is the focus of inflammatory pathologies. MPO could participate in the initiation and progression of periodontal disease.
Materials and Methods:
A total of 60 patients with healthy gingiva, gingivitis, periodontitis between age group of 20–55 years were selected. Group I - 20 Patients with healthy gingiva, Group II - 20 Patients with generalized gingivitis, Group III - 20 Patients with generalized chronic periodontitis, Group IV - 20 Patients of Group III after 1-month of scaling and root planning. The following parameters were recorded: Gingival index, plaque index, bleeding on probing index, probing pocket depth, clinical attachment level, salivary MPO levels. All the parameters were then statistically analyzed.
Results:
The mean MPO levels in Group I recorded was - 0.320 + 0.06, Group II was - 0.183 + 0.04, Group III was - 0.814 + 0.08 and Group IV was - 0.386 + 0.08 respectively. All these values were statistically significant when compared between the four groups (P < 0.05). A significantly elevated salivary MPO levels were found in subjects with chronic periodontitis as compared to the gingivitis group and the healthy group (P < 0.05). However, moderate but statistically significant increase in the MPO levels were observed in the gingivitis group as compared to the healthy group (P < 0.05). Furthermore, significant reduction in MPO levels were observed in Group IV after 1-month of nonsurgical periodontal therapy.
Conclusion:
The activities of MPO enzyme were significantly increased in the saliva of patients with periodontal disease in comparison to healthy individuals. Furthermore, nonsurgical periodontal therapy was found to be effective in improving clinical parameters and in reducing MPO levels. Salivary enzymes like MPO could be considered as a biochemical marker of periodontal disease activity.
Keywords: Ghronic periodontitis, gingivitis, myeloperoxidase, nonsurgical periodontal therapy, periodontal disease, spectrophotometer
INTRODUCTION
Chronic periodontitis is an inflammatory disease of the periodontium that extends beyond the gingiva and produces destruction of connective tissue attachment of the teeth.[1] The main functions of polymorphonuclear leukocytes (PMN’s) include adherence, chemotaxis, phagocytosis, and bactericidal activity. Patients with altered numbers or functionality of circulating PMN's tend to have severe periodontal disease.[2]
Myeloperoxidase (MPO), the most abundant protein in neutrophils is the focus of inflammatory pathologies.[1] The amount of MPO is 3 times higher in PMN cells than in monocytes, and MPO represents up to 5% of the total protein of PMN cells.[3] MPO is produced by PMNs and its concentration increases in the gingival fluid during inflammation, and that increase may be reflected in whole saliva too.[4] Association of MPO is found to be related to progression episodes and treatment responses in patients with chronic periodontitis.[4,5]
Traditional periodontal diagnosis involves measures of probing depth (PD), gingival recession, probing attachment level using graduated periodontal probe. These are indirect measures of bone loss. Biochemical markers can detect inflammatory changes in a short period where as a longer period is required to detect measurable changes in bone density using radiographs. The analysis of these enzymes in salivary secretion and inflammatory markers in serum of periodontitis patients can contribute to the clarification of the pathogenesis and improvement of making a prompt diagnosis of the periodontal disease.[6]
Myeloperoxidase produces hypochlorous acid (HOCl) from hydrogen peroxide (H2O2) and chloride anion (Cl−) (or the equivalent from a nonchlorine halide) during the neutrophil's respiratory burst. It requires heme as a co-factor. Furthermore, it oxidizes tyrosine to tyrosyl radical using H2O2 as an oxidizing agent.
Hypochlorous acid and tyrosyl radical are cytotoxic, so they are used by the neutrophil to kill bacteria and other pathogens.
The normal MPO levels vary between populations, with mean values between 20.310 and 59.25 units/ml. However, using digital spectrophotometer, it is possible to measure MPO levels as low as - 0.50 AU.
Thomas et al. indicated that MPO was responsible for a large portion of peroxidase-catalyzed reactions in mixed saliva. Increased number of MPO containing leukocytes indicated infection and inflammation of oral tissues.[7] Furthermore, Thomas et al. conducted a study to determine activity of MPO by spectrophotometer using 4-aminoantipyrine in serum and saliva of subjects with chronic periodontitis and he observed that activity of MPO is significantly increased in serum and saliva of patients with periodontal disease in relation to healthy individuals. Hence, the salivary enzymes like MPO could be considered as possible biochemical markers of the functional condition of periodontal tissues.[1]
Many researchers have assumed that activity of various enzymes in saliva, as biochemical markers for periodontal tissue damage, may be useful in the diagnosis, prognosis and evaluation of therapy effects in periodontal disease.[8,9,10,11,12,13]
Thus, in the present study, we compared and evaluated the levels of MPO in saliva samples of the patients with healthy gingiva, gingivitis, and in chronic periodontitis before and after scaling and root planning.
MATERIALS AND METHODS
Study population
A total of 60 subjects with good systemic health diagnosed as patients with healthy gingiva, gingivitis, chronic periodontitis were selected from the outpatient department, department of Periodontology, Subharti Dental College and Hospital, Meerut, Uttar Pradesh. Group I - 20 patients Healthy [Figure 1]. Group II - 20 patients Gingivitis [Figure 2]. Group III - 20 patients with generalized chronic periodontitis [Figure 3]. Group IV - 20 patients of Group III treated with scaling and root planning and recalled after 1-month [Figure 4].
Figure 1.
Group I - healthy
Figure 2.
Group II - gingivitis
Figure 3.
(a) Group III - chronic periodontitis (b) measurement of probing pocket depth by acrylic stent (c) orthopantomograph
Figure 4.
(a) Group IV - 1 month after non-surgical periodontal therapy (b) measurement of probing pocket depth with acrylic stent 1 month after nonsurgical periodontal therapy
Clinical and biochemical parameters
The following parameters were recorded: Gingival index (GI),[14] plaque index (PI),[15] gingival bleeding index (GBI),[16] PD (in mm), clinical attachment level (CAL), salivary MPO levels with spectrophotometer at 510 nm.
Procedure
Method of collection of saliva
Un stimulated salivary sample was collected. Patients were instructed not to consume any type of food within at least 1 h before saliva collection. Before any clinical examination approximately 2 ml of un-stimulated saliva was collected, stored and labeled. The saliva sample was collected in a sterile disposable plastic container, which was covered by a plastic lid.
Clinical attachment level for every tooth was assessed on all four surfaces of each tooth (i.e., mesiobuccal, buccal, distobuccal, and lingual/palatal). This was determined by measuring the distance from the base of the pocket to the fixed point on the tooth, that is, cemento-enamel junction. The mean of CAL was calculated for analysis.
Relative attachment level (RAL) was also measured by using acrylic stent only for standardization of values.
Data collection and clinical studies
A volume of 0.1 ml of saliva sample was taken and 0.5 ml sodium phosphate (pH - 6.1) was added which acted as buffer (i.e., - 0.1 ml saliva + 0.5 ml sodium phosphate). To that solution 0.5 ml prepared H2O2 was added. then, 0.5 ml 4-aminoantipyrene was added to the prepared mixture for spectrophotometer analysis (i.e., - 0.1 ml saliva + 0.5 ml sodium phosphate + 0.5 ml prepared H2O2 + 0.5 ml 4-aminoantipyrene). After addition of 4-aminoantipyrene, readings were recorded on spectrophotometer at 510 nm. Three recordings were obtained at the interval of 5 min (i.e., at 5 min, 10 min and 15 min) and the mean of these three values were recorded.
Statistical analysis
The data obtained were subjected to statistical analysis. Mean and standard deviation were estimated from the samples for each study group. Mean values were compared using one way analysis of variance. Unpaired t-test was used for intragroup comparison at same time, that is, baseline. And paired t-test was used for comparison between Group III and Group IV, that is, baseline and after 1-month of scaling and root planning. Other than this, Karl-Pearson's correlation coefficient test was used to measure the correlation of two indices with each other in every group. Also, percentage difference was calculated between Group III and Group IV. In the present study, P < 0.05 was considered as the level of significance.
RESULTS
In this study, results were obtained on the basis of GI, PI, GBI, PD, CAL, RAL.
The probing depth (in mm)
The mean PD in Group I recorded was 2.004 + 0.19, Group II was 2.408 + 0.28, Group III was 5.202 + 0.33 and Group IV was 3.861 + 0.38 respectively as shown in Table 1 and Graph I. When the mean PD score was compared between the Group I, Group II, Group III and Group IV, there was a statistical significant difference among the four groups [Table 2 and 3] at 5% level of significance.
Table 1.
Representing the mean and SD of different parameters in 4 groups
Graph I.
Comparison between the mean probing depth (in mm) for Group I, Group II, Group III and Group IV
Table 2.
Comparison between different pairs of groups for different parameters (unpaired t-test)
Table 3.
Comparison between Group III and Group IV (follow-up study) for different parameters (paired t-test)
The clinical attachment level (in mm)
The mean CAL in Group I recorded was 1.088 + 0.09, Group II was 1.261 + 0.34, Group III was 5.999 + 0.41 and Group IV was 4.518 + 0.48 respectively as shown in Table 1, Graph II. When the mean CAL was compared between the Group I, Group II, Group III and Group IV, there was a statistical significant difference among the four groups [Tables 2 and 3] at 5% level of significance.
Graph II.
Comparison between the mean clinical attachment level (in mm) for Group I, Group II, Group III and Group IV
The relative attachment level (in mm)
The mean RAL in Group III recorded was 11.168 + 1.37 and Group IV was 9.100 + 1.22 respectively as shown in Table 1. When the mean RAL was compared between the Group III and Group IV there was a statistical significant difference among the two groups [Tables 2 and 3] at 5% level of significance. Also, there was a difference between Group III and Group IV (P = 1.23373E-10) scores, which were statistically significant at 5% level of significance.
The MPO levels were measured in the saliva from all patients at baseline. It was measured by spectrophotometeric analysis at 510 nm. The mean MPO levels in Group I recorded was - 0.320 + 0.06, Group II was - 0.183 + 0.05, Group III was - 0.814 + 0.08 and Group IV was - 0.386 + 0.08 respectively as shown in Table 1 and Graph III. When the mean MPO was compared between the Group I, Group II, Group III and Group IV, there was a statistical significant difference among the four groups [Tables 2 and 3] at 5% level of significance.
Graph III.
Comparison between the mean myeloperoxidase levels (in AU) for Group I, Group II, Group III and Group IV
Average percentage difference was also evaluated between Group III and Group IV (follow-up study) to compare the percentage improvement in Group IV which was found to be - 71.47% for GI, 56.97% for PI, 68.70% for GBI, 36.05% for PD, 33.91% for CAL, 23.17% for RAL and 119.30% for MPO respectively as shown in Table 4.
Table 4.
Average % difference between Group III and Group IV (follow-up study)
DISCUSSION
In the present study, saliva was the choice of sample over GCF because it could be collected in a noninvasive and convenient manner. Saliva contains locally and systemically derived biomarkers of periodontal disorders and can therefore, be recommended as a rapid, noninvasive, on-site and patient specific diagnostic test.[8] However, collection of GCF from sulcus is technique sensitive and also continuous use of paper strips can also cause irritation to the gingiva thus due to irritation, GCF flow can increase which can lead to inaccurate findings.
Myeloperoxidase, a most abundant protein in neutrophils is the focus of inflammatory pathologies. It is used as a biomarker for human cardiovascular diseases like hypertension, etc., Its ability to catalyze a reaction between chloride and H2O2 to form HOCl is unique among mammalian enzymes. HOCl is a powerful antimicrobial agent and extremely reactive with biological molecules.[1,2,5] Thus, MPO has its role in host defense mechanism.[7] MPO is found in two classes of circulating white blood cells, the monocytes and PMN leukocytes (neutrophils). The amount of MPO is 3 times higher in PMN cells than in monocytes and MPO represents up to 5% of total protein of PMN cells. In resting PMN cells, the enzyme is stored in cytoplasmic secretory granules. Stimulated cells secrete MPO and H2O2 into phagolysosomes and the extracellular medium. MPO catalyzes the oxidation of chloride ion by H2O2 to produce the potent bactericidal agent HOCl. Antimicrobial activity of this system may be due to the direct reaction of HOCl with microbial cell components.[7]
Many researchers have assumed that activity of MPO in saliva, as biochemical markers for periodontal tissue damage, may be useful in the diagnosis, prognosis and evaluation of therapy effects in periodontal disease.[8,9,10]
Thus in the present study, we have selected MPO enzyme to further evaluate the significant role of this enzyme in relation to periodontal status as it is directly related to inflammatory pathologies as periodontitis.
There are many studies in which MPO levels have been measured and related to the periodontal diseases.[1,2,5,7,17,18,20,21,22] However most of them have used GCF as sample and used ELISA kit etc., for evaluation. However, in the present study we have collected saliva as a choice of sample to evaluate particularly its potential effect on this enzyme with periodontal diseases and also with newer equipment, digital spectrophotometer (Picogene digital spectrophotometer S. No. 195/11, manufactured by the Genetix Biotech Asia Pvt. Ltd., Germany). We have chosen this over other techniques as it provides accurate digital output as compared to others, and a digital graph is visible on the screen with different wavelengths. Also, it is highly specific and sensitive even up to the range of - 0.50–2.5. The added advantage of this technique is that it is less expensive as compared to other ELISA test or colorimetric test.
In a study conducted by Smith et al.[23] it was apparent from the results of the study that MPO appears in whole saliva of patients with gingivitis, which however decreased after the normal was resumed. This study was one of the earliest studies to determine the relationship between the MPO with periodontal disease. Later, Smith et al.[17] again revealed a positive relationship between GCF MPO and periodontitis, which decline following mechanical debridement.
In the present study, the mean MPO levels in Group I recorded was - 0.320 + 0.06, Group II was - 0.183 + 0.04, Group III was 0.814 + 0.08 and Group IV was 0.386 + 0.08 as shown in Table 1. When the mean MPO levels were compared between Group I, Group II, Group III and Group IV, there was a statistical significant difference among the four groups at 5% level of significance as shown in Table 1 (P = 1.70495E-60; P < 0.05). Also, a statistically significant difference was observed between Group I and Group II (P = 1.93978E-09; P < 0.05), Group I and Group III (P = 4.47355E-34; P < 0.05), Group I and Group IV (P = 7.35076E-28; P < 0.05), Group II and Group III (P = 2.86848E-29; P < 0.05), Group II and Group IV (P = 3.32976E-23; P < 0.05) and Group III and Group IV (P = 5.47338E-13; P < 0.05) scores respectively, at 5% level of significance. The periodontitis group (Group III) demonstrates a higher MPO levels (0.814 + 0.082) indicating risk for periodontal disease. This is in accordance with the findings of the earlier studies done by Thomas et al.[1] and Ramesh et al.[4]
Goncalves et al.[5] in a study estimated a mean MPO of 0.54 for periodontitis patients, which decreased up to 0.24 after nonsurgical periodontal therapy. In another study conducted by Cao et al.,[19] the highest levels of MPO in periodontitis patients were (mean) 0.75 + 0.07 as compared to healthy patients (mean) 0.18 + 0.06.
In this study, GI and PI and GBI have shown a positive correlation with the levels of MPO in chronic periodontitis group and gingivitis group as compared to the healthy controls suggesting that increase in the microbial colonization leads to the increased levels of MPO in saliva. Because as inflammation increases because of the presence of plaque and calculus in chronic periodontitis, then due to the defense mechanism neutrophils increase which in turn increases the MPO levels as MPO lies within the azurophilic granules of neutrophils.
Along with this, MPO levels increased with the increase in the PD, CAL and RAL in all the subjects of Group III, in which patients who had pocket depth of ≥5 mm (mean 5.20 ± 0.33) were included as compared to the healthy controls, who had healthy gingiva with no pocket depth. This is in accordance with the studies conducted by Cao et al.[19] and Smith et al.[20] in past where they found a positive correlation between MPO and chronic periodontitis patients with probing pocket depth of ≥4 mm (mean 5.02 ± 0.27). Also this increased levels of MPO was found to decrease with the reduction in pocket depth (mean 2.88 ± 1.03) after nonsurgical periodontal therapy which is in accordance with the results of the study conducted by Buchman et al.[24] It was also demonstrated by Hernandez et al.,[25] Bosnjak et al.[26] that MPO levels increased with periodontitis but also decreased after treatment of periodontal disease.
Thus the results obtained in the present study confirmed that all parameters GI, PI, GBI, PD, CAL were significantly associated with the chronic periodontitis and also salivary MPO levels were assessed and a strong correlation of MPO levels with the chronic periodontitis was established. Also with scaling and root planning all parameters were reduced to a significant level in Group IV and so did the MPO levels.
CONCLUSION
The results of the present study showed that the activities of MPO enzyme were significantly increased in the saliva of patients with periodontal disease in comparison to healthy individuals. Also, nonsurgical periodontal therapy was found to be effective in improving most clinical parameters and in reducing salivary enzymatic activity. Salivary enzymes like MPO could be considered as possible biochemical markers of periodontal disease and further to monitor the effectiveness of periodontal treatment.
Footnotes
Source of Support: Nil
Conflict of Interest: None declared.
REFERENCES
- 1.Thomas B, Alva M, Ramesh A, Kumari S. Comparative evaluation of myeloperoxidase levels in healthy individuals and subjects with chronic periodontitis – A biochemical study. J Soc Periodontists Implantologists Kerala. 2010;4:39–41. [Google Scholar]
- 2.Chang LY, Lin JC, Chang CW, Ho WH, Chen YT, Peng JL, et al. Inhibitory effects of safrole on phagocytosis, intracellular reactive oxygen species, and the activity of myeloperoxidase released by human polymorphonuclear leukocytes. J Periodontol. 2009;80:1330–7. doi: 10.1902/jop.2009.080202. [DOI] [PubMed] [Google Scholar]
- 3.Fabian TK, Fejerdy P. Saliva in health and disease and chemically biology of saliva. Wiley Encyclopedia Chem Biol. 2007;10:1–10. [Google Scholar]
- 4.Ramesh A, Thomas T, Anand A, Kumari S. Comparative evaluation of myeloperioxidase levels among systemically healthy individuals and patients with type 2 diabetes mellitus, with chronic periodontitis – Biochemical study. J Soc Periodontists Implantologists Kerala. 2010;4:16–9. [Google Scholar]
- 5.Gonçalves D, Correa FO, Khalil NM, de Faria Oliveira OM, Orrico SR. The effect of non-surgical periodontal therapy on peroxidase activity in diabetic patients: A case-control pilot study. J Clin Periodontol. 2008;35:799–806. doi: 10.1111/j.1600-051X.2008.01289.x. [DOI] [PubMed] [Google Scholar]
- 6.Yamalik N, Caglayan F, Kilinç K, Kilinç A, Tümer C. The importance of data presentation regarding gingival crevicular fluid myeloperoxidase and elastase-like activity in periodontal disease and health status. J Periodontol. 2000;71:460–7. doi: 10.1902/jop.2000.71.3.460. [DOI] [PubMed] [Google Scholar]
- 7.Thomas EL, Jefferson MM, Joyner RE, Cook GS, King CC. Leukocyte myeloperoxidase and salivary lactoperoxidase: Identification and quantitation in human mixed saliva. J Dent Res. 1994;73:544–55. doi: 10.1177/00220345940730021001. [DOI] [PubMed] [Google Scholar]
- 8.Rezaei A, Sariri R. Salivary diagnosis of periodontitis status: A review. Pharm Online. 2011;1:428–44. [Google Scholar]
- 9.Zhang L, Henson BS, Camargo PM, Wong DT. The clinical value of salivary biomarkers for periodontal disease. Periodontol 2000. 2009;51:25–37. doi: 10.1111/j.1600-0757.2009.00315.x. [DOI] [PubMed] [Google Scholar]
- 10.Taba M, Jr, Kinney J, Kim AS, Giannobile WV. Diagnostic biomarkers for oral and periodontal diseases. Dent Clin North Am. 2005;49:551–71. doi: 10.1016/j.cden.2005.03.009. vi. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Todorovic T, Dozic I, Vicente-Barrero M, Ljuskovic B, Pejovic J, Marjanovic M, et al. Salivary enzymes and periodontal disease. Med Oral Patol Oral Cir Bucal. 2006;11:E115–9. [PubMed] [Google Scholar]
- 12.Jentsch H, Sievert Y, Göcke R. Lactoferrin and other markers from gingival crevicular fluid and saliva before and after periodontal treatment. J Clin Periodontol. 2004;31:511–4. doi: 10.1111/j.1600-051X.2004.00512.x. [DOI] [PubMed] [Google Scholar]
- 13.Kumar R, Sharma G. Salivary alkaline phosphatase levels as diagnostic marker of periodontal disease. J Int Oral Health. 2011;3:81–5. [Google Scholar]
- 14.Loe H, Silness J. Periodontal disease in pregnancy II. Prevalance and severity. Acta Odontol Scand. 1963;21:533. doi: 10.3109/00016356309011240. [DOI] [PubMed] [Google Scholar]
- 15.Silness J, Loe H. Periodontal disease in pregnancy. II. Correlation between oral hygiene and periodontal condtion. Acta Odontol Scand. 1964;22:121–35. doi: 10.3109/00016356408993968. [DOI] [PubMed] [Google Scholar]
- 16.Ainamo J, Bay I. Problems and proposals for recording gingivitis and plaque. Int Dent J. 1975;25:229–35. [PubMed] [Google Scholar]
- 17.Smith QT, Hinrichs JE, Melnyk RS. Gingival crevicular fluid myeloperoxidase at periodontitis sites. J Periodontal Res. 1986;21:45–55. doi: 10.1111/j.1600-0765.1986.tb01437.x. [DOI] [PubMed] [Google Scholar]
- 18.Wolff LF, Smith QT, Snyder WK, Bedrick JA, Liljemark WF, Aeppli DA, et al. Relationship between lactate dehydrogenase and myeloperoxidase levels in human gingival crevicular fluid and clinical and microbial measurements. J Clin Periodontol. 1988;15:110–5. doi: 10.1111/j.1600-051x.1988.tb01003.x. [DOI] [PubMed] [Google Scholar]
- 19.Cao CF, Smith QT. Crevicular fluid myeloperoxidase at healthy, gingivitis and periodontitis sites. J Clin Periodontol. 1989;16:17–20. doi: 10.1111/j.1600-051x.1989.tb01606.x. [DOI] [PubMed] [Google Scholar]
- 20.Smith QT, Au GS, Freese PL, Osborn JB, Stoltenberg JL. Five parameters of gingival crevicular fluid from eight surfaces in periodontal health and disease. J Periodontal Res. 1992;27:466–75. doi: 10.1111/j.1600-0765.1992.tb01819.x. [DOI] [PubMed] [Google Scholar]
- 21.Puklo M, Guentsch A, Hiemstra PS, Eick S, Potempa J. Analysis of neutrophil-derived antimicrobial peptides in gingival crevicular fluid suggests importance of cathelicidin LL-37 in the innate immune response against periodontogenic bacteria. Oral Microbiol Immunol. 2008;23:328–35. doi: 10.1111/j.1399-302X.2008.00433.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Trombone AP, Ferreira SB, Jr, Raimundo FM, de Moura KC, Avila-Campos MJ, Silva JS, et al. Experimental periodontitis in mice selected for maximal or minimal inflammatory reactions: Increased inflammatory immune responsiveness drives increased alveolar bone loss without enhancing the control of periodontal infection. J Periodontal Res. 2009;44:443–51. doi: 10.1111/j.1600-0765.2008.01133.x. [DOI] [PubMed] [Google Scholar]
- 23.Smith AJ, Smith G, Basu MK, Walsh TF. Changes in salivary peroxidase activity observed during experimentally-induced gingivitis. J Clin Periodontol. 1984;11:373–8. doi: 10.1111/j.1600-051x.1984.tb01335.x. [DOI] [PubMed] [Google Scholar]
- 24.Buchmann R, Hasilik A, Nunn ME, Van Dyke TE, Lange DE. PMN responses in chronic periodontal disease: Evaluation by gingival crevicular fluid enzymes and elastase-alpha-1-proteinase inhibitor complex. J Clin Periodontol. 2002;29:563–72. doi: 10.1034/j.1600-051x.2002.290613.x. [DOI] [PubMed] [Google Scholar]
- 25.Hernández M, Gamonal J, Tervahartiala T, Mäntylä P, Rivera O, Dezerega A, et al. Associations between matrix metalloproteinase-8 and -14 and myeloperoxidase in gingival crevicular fluid from subjects with progressive chronic periodontitis: A longitudinal study. J Periodontol. 2010;81:1644–52. doi: 10.1902/jop.2010.100196. [DOI] [PubMed] [Google Scholar]
- 26.Bosnjak A, Boras VV, Lukac J, Brozovic S. Whole saliva levels of some inflammatory mediators in patients with previous evidence of periodontitis: A pilot study. Acta Stomatol Croat. 2009;43:24–33. [Google Scholar]
