JOM 2000-2297 Journal of Oral Microbiology, Vol. 1, August 2009, pp. 1–9 sici xxx PUBMED Abbreviation LauraLakio JohannaAntinheimo SusannaPaju KåreBuhlin

*Kåre Buhlin Institute of Dentistry University of HelsinkiP.O. Box 6300014HelsinkiFinlandnd fax: +358 9 19125194

PirkkoJ.Pussinen GeorgAlfthan Institute of Dentistry University of Helsinki

HelsinkiFinland

Department of Oral and Maxillofacial Diseases Helsinki University Central Hospital

HelsinkiFinland

Biomarker Laboratory, Department of Health and Functional Capacity National Public Health Institute

HelsinkiFinland

Journal of Oral Microbiology ORIGINAL ARTICLE Tracking of plasma antibodies Background: Plasma antibody measurements of antibody levels to periodontal pathogens may be used to support diagnosis, disease activity, classification, and prognosis of periodontitis. Objective: The aim of this study was to investigate the long-term stability of plasma antibody levels against Aggregatibacter actinomycetemcomitans and Porphyromonas gingivalis. Design: Plasma immunoglobulin G (IgG) antibody levels against the pathogens were analyzed annually during 15 years from 21 voluntary subjects, whose periodontal status was not known at the point of selection. The total number of plasma samples was 315. In connection of the last sampling, the clinical and radiographic periodontal status was examined. Pooled bacterial samples from periodontal pockets, as well as salivary samples were collected for A. actinomycetemcomitans and P. gingivalis detection, and antibody determinations, respectively. According to the clinical status, six subjects had periodontitis, whereas 15 did not. Results: Plasma IgG-class antibody levels to periodontal pathogens remained extremely stable during the 15-year period and no significant (p>0.05) intra-individual variations were observed. Retrospectively, the average plasma IgG antibody levels against A. actinomycetemcomitans and P. gingivalis were 1.6–2.3 (p<0.05) and 1.4–1.7 (p<0.05) fold higher in the subjects with periodontitis than those without, respectively, during the whole 15-year tracking. As expected, at the time of the periodontal examination the plasma and salivary IgG antibody levels were associated both with periodontitis and bacterium-positivity. Conclusions: Plasma IgG levels against A. actinomycetemcomitans and P. gingivalis are extremely stable during 15 years both in subjects with and without periodontitis. Aggregatibacter actinomycetemcomitans longitudinal studies periodontitis plasma oral infections Porphyromonas gingivalis saliva Major periodontal pathogens include Aggregatibacter actinomycetemcomitans and Porphyromonas gingivalis. The association between P. gingivalis and periodontitis is stronger even though a carrier-state of the species may exist 1. A. actinomycetemcomitans has a dual role in the oral flora: it has been associated with aggressive forms of periodontitis 2, but the bacterium can also be found in the oral cavity of healthy individuals 3. The bacterial load in periodontal pockets leads to the activation of host-bacterial interactions. To establish an infection, the pathogen must overcome numerous surface barriers, such as epithelium and mucus, and also the innate immunity and adaptive immunity 4. Periodontal pathogens stimulate B cell activation mainly through classical antigen-specific immune response. A. actinomycetemcomitans and P. gingivalis are capable of inducing high antibody response that can in fact exceed those found in many other pathological bacterial infections 5.Plasma and saliva antibody measurements have been used to diagnose periodontitis, estimate its activity, classification and prognosis, and success of treatment 5. Several studies have reported elevated levels of plasma immunoglobulin G (IgG) against periodontal pathogens especially in patients with aggressive periodontitis 6789. Immunoglobulin A (IgA) is the predominant immunoglobulin in saliva, and salivary IgA plays a role in the local immune defense system. In addition to the plasma-derived IgG and IgA, certain subclasses of IgG and IgA can be produced locally in the periodontal pockets 10. Antibody production, especially that of IgG and IgA, is considered to have a protective role in the pathogenesis of periodontitis 1112. It has also been suggested that antibodies may not necessarily need to decrease periodontal pathogen loads in order to exert a protective effect, but they may neutralize toxins and proteolytic enzymes as well as promote opsonization, and complement activation 11.The stability of IgG-class antibody levels has been investigated in a few follow-up studies. The follow-up time in those studies has varied from two to three years 1314. IgG-class antibody levels showed long-term stability; no decrease has been reported after periodontal treatment 614 or treatment has lowered the IgG levels only temporarily 15.Earlier studies have mainly focused on the relationship between antibody levels and severity of periodontitis or between bacterial presence and their homologous antibodies. The aim in this study was therefore to investigate the long-term stability of plasma antibody levels against A. actinomycetemcomitans and P. gingivalis in plasma samples analyzed annually from the same individuals during 15 years. This report is based on the 15-year collection of blood samples from a rural population in Leppävirta, Savolax, Finland, whose selenium status was followed by annual blood sampling. The original sample included 26 men and 19 women in 1985 16. The local public health-care center recruited subjects. Volunteers for the periodontal examination were recruited by a questionnaire, and 9 men and 12 women participated. During the 15 years, the subjects had had dental check-ups by their general practitioners and none of them had previously been treated by a periodontologist. However, they were instructed to seek periodontal treatment after the examination if necessary. The characteristics of the participants at the end of the study are shown in Table 1. All the participating subjects gave their informed consent. Periodontitis Periodontally healthy Variable n (%) n (%) p-Value^{a a}Mann–Whitney test or Chi-square test. ^bCPITN = communal periodontal index in treatment needs. Tracking of plasma antibodies against Aggregatibacter actinomycetemcomitans and Porphyromonas gingivalis during 15 years. ^cRadiological analyses. Note: ns = non-significant. Gender  Men 1 (17) 8 (53) ns  Women 5 (83) 7 (47) ns General health  Elevated blood pressure 3 (50) 2 (13) ns  Cardiovascular diseases 2 (33) 0 ns  Diabetes (type II) 0 1 (7) ns  Osteoporosis 0 1 (7) ns  Current smoking 0 3 (20) ns Mean±SD Mean±SD Age 66±5 59±13 0.302 Periodontal examination  Total number of teeth 19.5±4.1 15.7±10.8 0.569  Visible plaque (% of sites) 13.33±9.2 11.2±9.6 0.558  Bleeding on probing (% of sites) 36.8±9.5 24.6±17.2 0.112  Number of teeth with periodontal pockets (≥5 mm) 2.0±2.5 0 0.000  Periodontal pockets (% of sites) 1.8±2.3 0 0.000  Clinical attachment loss (mm) 4.17±4.54 0 0.000  Highest CPITN value^b 3.8±0.4 2.4±0.5 0.000  Total number of angular bony defects^c 1.5±2.0 0.3±0.5 0.213 The subjects were examined with respect to visible plaque and bleeding on probing 17, probing depth (periodontal pockets ≥5 mm), and clinical attachment loss at six sites per tooth, at all present teeth (Table 1). One dentist (JA) performed all the clinical examinations. WHO probe (LM Dental LM 550B Si, LM-instruments Oy/Ab, Pargas, Finland) was used in the examinations. Subjects were categorized into those with and without periodontitis according to the following criteria: subjects with periodontitis had at least one site with a periodontal pocket deeper than 5 mm and clinical attachment loss (see Table 1).Periodontal status and possible infectious foci of periodontium and/or alveolar bone were detected from analog panoramic tomographs. The number of peri-apical lesions, angular bony defects (they were at least to the middle third of the root length), furcation lesions in molars, and pericoronitis in third molars were recorded. Plasma samples were collected once a year for 15 years from the 21 participants and stored at −20°C. The following samples were collected at the end of the 15-year period.Paraffin-stimulated salivary samples were collected for five minutes. The salivary sample was divided into two aliquots, of which 2 ml was centrifuged immediately at 12,000 rpm for five minutes and frozen on dry ice. The frozen samples were brought to the research laboratory at the Institute of Dentistry, University of Helsinki, Helsinki, within 48 hours and stored at −70°C. From the 21 participants in the study, pooled bacterial samples from the 14 most inflamed periodontal pockets in each subject were collected with sterile curettes at the end of the study. The samples were transported to the laboratory in vials containing 2 ml of viability preserving medium no. III (VMGA III) medium, and cultured. A. actinomycetemcomitans and P. gingivalis were identified as previously described 1819. Briefly, for A. actinomycetemcomitans, 100-µl aliquots of undiluted and 10⁻² dilutions of samples were inoculated on Trypic soy bacitracin vancomycin (TSBV)-agar plates and incubated in 5% CO₂ in air at 36°C for three days. A. actinomycetemcomitans grew as typical adherent colonies, which were catalase-positive. For P. gingivalis, 100-µl aliquots of 10⁻⁴ and 10⁻⁵ were inoculated on Brucella agar plates and incubated in anaerobic jars at 37°C for 10 days. P. gingivalis grew as dark-pigmented colonies, which had positive trypsin-like enzyme activity as detected with carbo-benzoxy-L-arginine-7-amino-4-methylcoumarin amide HCl (CAAM)-reagent.Bacterial DNA was isolated from the VMGA III medium according to the manufacturer's instructions using Chelex 100 resin, and A. actinomycetemcomitans and P. gingivalis were detected by PCR as reported earlier 2021. In every series of PCR, chromosomal DNA extracted from A. actinomycetemcomitans (ATCC43718) and P. gingivalis (W50) strains served as positive controls, and water served as a negative control.Plasma IgG and salivary IgG and IgA antibodies against A. actinomycetemcomitans and P. gingivalis were detected by multiserotype-ELISA as previously described 6. The strains used as formalin-killed whole cell antigens were ATCC29523, ATCC43781, ATCC33384, IDH781, IDH1708, and C59 representing A. actinomycetemcomitans serotypes a, b, c, d, e, and a non-serotypeable strain, respectively, and ATCC33277, W50, and OMGS434 representing P. gingivalis serotypes a, b, and c, respectively. Plasma or saliva diluted in antibody buffer (PBS containing 0.05% Tween 20 and 0.5% bovine plasma albumin) was incubated at room temperature for two hours. For the determinations, four dilutions of each sample were used in duplicate. The dilutions were as follows: plasma samples for the detection of A. actinomycetemcomitans IgG 1:500, 1:1,500, 1:4,500, and 1:13,500 and for the detection of P. gingivalis IgG 1:100, 1:400, 1:1,600, and 1:6,400, and salivary samples for the detection of either A. actinomycetemcomitans or P. gingivalis IgG and IgA were 1:3.3, 1:10, 1:30, and 1:90. The results were calculated as Area under the plasma concentration time curve (AUC) from the dilution curves. The inter-assay coefficients for variation as calculated from values of the reference serum applied on each plate were 6.3 and 5.5% for A. actinomycetemcomitans and 5.1 and 4.7% for P. gingivalis IgA and IgG, respectively. Data analysis was performed using the SPSS for Windows, version 13.0 (SPSS Inc, Chicago, IL, USA). Differences in the clinical characteristics and antibody levels between the two study groups were analyzed by the Mann − Whitney U test or Chi-square test. Interactions between parameters were examined using Pearson Correlation. The statistical significance of the intra-individual differences in annual plasma IgG levels was analyzed by the Kruskall–Wallis test. P-values ≤0.05 were considered as statistically significant. Table 1 provides a summary of the demographic and clinical data of the subject sample. According to our classification criteria, the subjects with periodontitis at the time of the clinical examination had higher number of deep (≥5 mm) periodontal pockets, attachment loss levels, and community periodontal index of treatment needs (CPITN) value compared to those without periodontitis. No differences were found in the bleeding on probing and visible plaque or in the number of angular bony defects detected from the radiographic pictures between the two groups. There were no statistical differences in the general health and current smoking between the subjects with (n=6) and without (n=15) periodontitis. Mean plasma and salivary IgG antibody and salivary IgA antibody levels against A. actinomycetemcomitans and P. gingivalis in the periodontitis and non-periodontitis groups at the end of the study are shown in Table 2. Mean salivary IgG antibody levels against A. actinomycetemcomitans and both salivary and plasma IgG levels against P. gingivalis were significantly higher in subjects with periodontitis compared to the levels in the periodontally healthy subjects. There were no differences in the salivary IgA antibody levels against A. actinomycetemcomitans and P. gingivalis between the groups. Variable Periodontitis Periodontally healthy p-Value^a Mean±SD Mean±SD ^aMann–Whitney test. ^bEU = ELISA units. Plasma antibody levels (EU)^b A. actinomycetemcomitans IgG 20.51±12.76 11.06±5.70 0.066 P. gingivalis IgG 20.35±6.76 11.94±4.26 0.008 Salivary antibody levels (EU) A. actinomycetemcomitans IgG 0.63±0.58 0.14±0.16 0.045 A. actinomycetemcomitans IgA 1.11±0.88 0.59±0.50 0.205 P. gingivalis IgG 0.44±0.39 0.03±0.02 0.001 P. gingivalis IgA 0.85±0.76 0.46±0.37 0.235 The mean annual plasma antibody IgG levels against A. actinomycetemcomitans and P. gingivalis during the 15-year period in subjects with and without periodontitis are shown in Figure 1. Plasma IgG levels against A. actinomycetemcomitans were 1.6–2.3 fold higher in subjects with periodontitis than in those without (p<0.05). Mean IgG levels against P. gingivalis were 1.4–1.7 fold higher in subjects with periodontitis than in periodontally healthy subjects (p<0.05). The intra-individual variation in plasma IgG antibodies against P. gingivalis (Figure 2a) and A. actinomycetemcomitans (Figure 2b) during the 15 years are shown for all subjects. The individual variation during the study was not statistically significant in any of the subjects.

Plasma immunoglobulin G (IgG) levels against A. actinomycetemcomitans and P. gingivalis levels during 15 years. The classification into those with and without periodontitis was done in the connection of the last sampling. The A. actinomycetemcomitans levels (ELISA units, mean±SEM) of periodontitis subjects (n=6) (♦) and periodontally healthy subjects (n=15) (⋄), P. gingivalis levels (ELISA units, mean±SEM) in periodontitis subjects (•) and periodontally healthy subjects (○). Statistically significant differences are marked with asterisk (p<0.05).

Individual variation in plasma immunoglobulin G (IgG) levels (ELISA units) against P. gingivalis (a) and A. actinomycetemcomitans (b) during 15 years. The box plot columns show antibody levels in each year (○), mean antibody levels (•), and standard deviation (vertical lines). One box plot column represents one subject. Subjects marked with # are those who were PCR-positive for P. gingivalis (a) and A. actinomycetemcomitans (b) at the end of the study.

Mean plasma and salivary IgG levels were associated with the bacteria detected by PCR (Table 3). IgG levels against A. actinomycetemcomitans and P. gingivalis in both plasma and saliva were significantly higher in subjects positive for A. actinomycetemcomitans and P. gingivalis compared to the subjects negative for the bacteria, respectively. Salivary IgA levels against A. actinomycetemcomitans and P. gingivalis were not significantly elevated in PCR-positive subjects for either of the bacteria. Of the two A. actinomycetemcomitans PCR-positive subjects one was culture-positive. Seven subjects were PCR-positive for P. gingivalis, and five of them were culture-positive. PCR negative PCR-positive Mean±SD Mean±SD p-Value^a Antibodies against A. actinomycetemcomitans Negative for A. actinomycetemcomitans Positive for A. actinomycetemcomitans ^aMann–Whitney U test. ^bSubjects negative for PCR detection of. A. actinomycetemcomitans. ^cSubjects positive for PCR detection of A. actinomycetemcomitans. ^dSubjects negative for PCR detection of P. gingivalis. ^eSubjects positive for PCR detection of P. gingivalis. n=19^b n=2^c IgG in plasma 11.45±5.70 35.72±1.65 0.023 IgG in saliva 5.05±5.59 21.29±1.84 0.023 IgA in saliva 16.72±10.05 27.73±6.47 0.093 Antibodies against P. gingivalis Negative for P. gingivalis Positive for P. gingivalis n=14^d n=7^e IgG in plasma 11.89±4.24 19.24±7.06 0.014 IgG in saliva 0.03±0.02 0.38±0.39 0.007 IgA in saliva 0.48±0.38 0.76±0.72 0.456 Correlation coefficients between the periodontal status and plasma, and salivary IgG and IgA antibody levels against A. actinomycetemcomitans and P. gingivalis at the end of the study are shown in Table 4. The number of angular bony defects correlated with IgG antibody levels against P. gingivalis in plasma and saliva and IgA antibody levels against A. actinomycetemcomitans in saliva. The number of deepened periodontal pockets (≥5 mm) correlated with IgG antibody levels against P. gingivalis in both plasma and saliva, IgA antibody levels against P. gingivalis in saliva, and IgG and IgA antibody levels against A. actinomycetemcomitans in saliva. Mean clinical attachment loss correlated with IgG and IgA antibody levels against P. gingivalis and A. actinomycetemcomitans in both plasma and saliva. The highest CPITN value correlated with IgG levels against P. gingivalis and A. actinomycetemcomitans in both plasma and saliva. Pg IgG _{in plasma} Aa IgG _{in plasma} Pg IgA ^{in saliva} Pg IgG _{in saliva} Aa IgA _{in saliva} Aa IgG _{in saliva} Highest CPI value Mean attachment loss (mm) Periodontal pockets^a Vertical bone loss (number of pockets) Bleeding on probing (%) ^aNumber of sites with deepened (≥5 mm) periodontal pockets. ^bCPI = clinical plaque index. ^cELISA units per stimulated salivary flow rate. *Statistically significant (p<0.05). Bleeding on probing (%) 0.164 0.225 0.064 0.262 0.125 0.210 0.233 0.258 0.170 0.050 1 Vertical bone loss (number of pockets) 0.589* 0.264 0.660* 0.623* 0.633* 0.432 0.381 0.451 0.901 1 Periodontal pockets^a 0.564* 0.315 0.515* 0.642* 0.548* 0.438* 0.535* 0.380 1 Mean attachment loss (mm) 0.637* 0.637* 0.660* 0.895* 0.581* 0.798* 0.598* 1 Highest CPI^b value 0.628* 0.526* 0.224 0.663* 0.236 0.498* 1 Aa IgG^c _{in saliva} 0.664* 0.881* 0.719* 0.895* 0.713* 1 Aa IgA^c _{in saliva} 0.513* 0.497* 0.938* 0.706* 1 Pg IgG^c _{in saliva} 0.783* 0.729* 0.732* 1 Pg IgA^c _{in saliva} 0.452* 0.504* 1 Aa IgG^c _{in plasma} 0.493* 1 Pg IgG^c _{in plasma} 1 The present study is a retrospective 15-year tracking of plasma antibody levels against A. actinomycetemcomitans and P. gingivalis. Throughout the whole time frame, the mean plasma antibody levels remained extremely stable, and only minor, non-significant individual variation was seen. Furthermore, our data showed elevated mean antibody levels against both A. actinomycetemcomitans and P. gingivalis in subjects, who were diagnosed as having periodontitis at the end of the 15-year time period, compared to the periodontally healthy.The main result in the present study was the remarkable stability of plasma IgG levels in both the periodontitis and periodontally healthy subjects, even though the periodontal classification was done at the end of the study. The limitation of the study is that we did not have the dental history of the subjects who participated in the study, and it is possible that they had had active periodontitis at some point during the 15 years. This is, however, not likely since at the time of the dental examination, none of them were currently in periodontal treatment or in the maintenance care. The subjects had got occasional periodontal treatment by their general practitioners, but none of them had sought for treatment by a specialized periodontologist.The antibody detection method we use in our laboratory has been published earlier 22. Antigens in the assay represent A. actinomycetemcomitans serotypes a–e and a non-serotypeable strain, and P. gingivalis serotypes a–c 6. When the IgG-class antibody levels to pathogens are summed up, the assay has a specificity and sensitivity of 90 and 71%, respectively, for finding periodontitis by measuring antibody levels from plasma samples 6. Since the aim of the present study, however, was not to validate further the multiserotype-ELISA, the antibody levels were not summed up.Our findings are consistent with previous studies. Papapanou and co-workers demonstrated an overall stability of plasma IgG-class antibody titers to 19 periodontal bacteria over a 30-month period for both periodontitis patients and periodontally intact control subjects 14. Periodontal treatment among those with the disease did not affect the plasma antibody levels. Furthermore, Dye and co-workers (2009) reported that serum IgG titers to selected periodontal species are feasible in epidemiologic studies using a combined serologic/demographic approach 22. Several other studies have shown higher plasma IgG levels against specific periodontal pathogens in patients with periodontitis than in periodontally healthy individuals 592425. It has been suggested that elevated antibody levels against A. actinomycetemcomitans and P. gingivalis exert a protective role in the progression of periodontitis 11. In a 36-month follow-up study, the patients who had low homologous plasma antibody levels together with cultivable A. actinomycetemcomitans and P. gingivalis, showed a positive predictive value for periodontitis disease recurrence 11. Our retrospective study, however, does not suggest a protective role for the plasma IgG-class antibody levels.Cultivable A. actinomycetemcomitans and P. gingivalis can serve as markers for destructive periodontal disease in adult subjects 26. In the present study, we were able to detect two A. actinomycetemcomitans-positive subjects and seven P. gingivalis-positive subjects by PCR. From those subjects, bacteria could be cultivated in one and five cases, respectively. Plasma and salivary IgG levels were elevated in the P. gingivalis-positive subjects as detected with both PCR method and cultivation. In the A. actinomycetemcomitans-positive subjects, the plasma and salivary IgG levels were elevated in the PCR-positive group. This can be explained by the small number of A. actinomycetemcomitans-positive subjects in the population.Plasma and salivary IgG levels correlated with each other. At the year of periodontal examination, both plasma and salivary antibody levels against P. gingivalis were significantly higher in the periodontitis group than in the periodontally healthy group. Concerning A. actinomycetemcomitans only the salivary IgG levels were elevated in the periodontitis group. A trend was seen also for plasma IgG against A. actinomycetemcomitans to be elevated in the periodontitis group even though not statistically significant. The small study population may explain this; from 28 subjects participating originally in the study annually during the 15 years, only 21 attended the clinical examination. In addition, all affected subjects had chronic periodontitis, that is more commonly associated with the presence of P. gingivalis than A. actinomycetemcomitans 27. Although IgA is the predominant immunoglobulin in saliva, our study showed elevated salivary IgG levels against both A. actinomycetemcomitans and P. gingivalis in periodontitis. In the whole saliva samples, the total immunoglobulin originates both from crevicular fluid and exocrine secretion. Therefore, there is some discrepancy compared to previous studies, in which salivary IgA have been measured in subjects with periodontitis.Hägewald and co-workers reported that a group of patients with aggressive periodontitis had lower total IgA levels than periodontally healthy subjects 28, whereas Henskens and associates found no difference in saliva IgA concentrations between patients with chronic periodontitis and periodontally healthy subjects 29. The present study supports the latter observation showing no differences in the IgA levels of periodontitis and periodontally healthy groups.Overall, plasma and salivary IgG and IgA levels against A. actinomycetemcomitans and P. gingivalis correlated significantly with periodontal status. There was a positive correlation with the number of sites with deepened periodontal pockets, mean attachment loss, and angular bony defects detected from the radiographic pictures. Only three of the 21 participants in the present study were current smokers, and all of them belonged to the periodontally healthy group. Smoking is a strong risk factor for periodontal diseases and may have an impact also on immunoglobulin levels 81230, but in this study the number of current smokers was too small to study the effect of smoking on the antibody production.In conclusion, the mean plasma IgG levels against A. actinomycetemcomitans and P. gingivalis were stable during the entire 15-year period. The mean plasma antibody levels were also correlated with the periodontal status, the levels were significantly higher in subjects with periodontitis compared to the periodontally healthy. In this study population, periodontitis was moderate and the number of subjects with periodontitis was small. Therefore, further studies are needed to confirm the observations among patients with severe periodontitis. The authors declare that they have no conflict of interest. This study was supported by the Academy of Finland (#118391 for PJP). Ms Tiina Karvonen is acknowledged for excellent technical assistance. The study was financially supported by the Academy of Finland (#118391 for PJP) and Finnish dental society Apollonia (for LL). KönönenEPajuSPussinenPJHyvönenMDi TellaPSuominen-TaipaleL Population-based study of salivary carriage of periodontal pathogens in adults J Clin Microbiol 20078 244651 SlotsJReynoldsHSGencoRJ. Actinobacillus actinomycetemcomitans in human periodontal disease: a cross-sectional microbiological investigation Infect Immun 198029 101320 ZambonJJ. Actinobacillus actinomycetemcomitans in human periodontal disease J Clin Periodontol 198512 120 DelvesPJRoittIM. The immune system. Second of two parts New Engl J Med 2000343 10817 KinaneDFMooneyJEbersoleJL. Humoral immune response to Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in periodontal disease Periodontol 2000 199920 289340 PussinenPJVilkuna-RautiainenTAlfthanGMattilaKAsikainenS. Multiserotype enzyme-linked immunosorbent assay as a diagnostic aid for periodontitis in large-scale studies J Clin Microbiol 200240 5128 ChungHYLuHCChenWLLuCTYangYHTsaiCC. Immunoglobulin G profiles in different forms of periodontitis J Periodontal Res 200338 4716 GraswinckelJEvan der VeldenUvan WinkelhoffAJHoekFJLoosBG. Plasma antibody levels in periodontitis patients and controls J Clin Periodontol 200431 5628 WangDKawashimaYNagasawaTTakeuchiYKojimaTUmedaM Elevated serum IgG titer and avidity to Actinobacillus actinomycetemcomitans serotype c in Japanese periodontitis patients Oral Microbiol Immunol 200520 1729 LundqvistCBaranovVTeglundSHammarströmSHammarströmML. Cytokine profile and ultrastructure of intraepithelial gamma delta T cells in chronically inflamed human gingiva suggest a cytotoxic effector function J Immunol 1994153 230212 RamsTEListgartenMASlotsJ. Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis subgingival presence, species-specific serum immunoglobulin G antibody levels, and periodontitis disease recurrence J Periodontal Res 200641 22834 KinaneDFMark BartoldP. Clinical relevance of the host responses of periodontitis Periodontol 2000 200743 27893 EhmkeBBeiklerTHaubitzIKarchHFlemmigTF. Multifactorial assessment of predictors for prevention of periodontal disease progression Clin Oral Investig 20037 21721 PapapanouPNNeiderudAMDisickELallaEMillerGCDahlénG. Longitudinal stability of serum immunoglobulin G responses to periodontal bacteria J Clin Periodontol 200431 98590 EbersoleJLCappelliDSteffenMJWillmannDEO'DellDS. Host response assessment in recurring periodontitis J Clin Periodontol 199623 25862 VaroPAlfthanGEkholmPAroAKoivistoinenP. Selenium intake and serum selenium in Finland: effects of soil fertilization with selenium Am J Clin Nut 198848 3249 AinamoJBayI. Problems and proposals for recording gingivitis and plaque Int Dent J 197525 22935 AsikainenSAlaluusuaSKariKKleemola-KujalaE. Subgingival microflora and periodontal conditions in healthy teenagers J Periodontol 198657 5059 SaarelaMStuckiAMvon Troil-LindénBAlaluusuaSJousimies-SomerHAsikainenS. Intra- and inter-individual comparison of Porphyromonas gingivalis genotypes FEMS Immunol Med Microbiol 19936 99102 AshimotoAChenCBakkerISlotsJ. Polymerase chain reaction detection of 8 putative periodontal pathogens in subgingival plaque of gingivitis and advanced periodontitis lesions Oral Microbiol Immunol 199611 26673 MättöJSaarelaMAlaluusuaSOjaVJousimies-SomerHAsikainenS. Detection of Porphyromonas gingivalis from saliva by PCR by using a simple sample-processing method J Clin Microbiol 199836 15760 Vilkuna-RautiainenTPussinenPJMattilaKVesanenMAhmanHDoganB Antigenically diverse reference strains and autologous strains of Actinobacillus actinomycetemcomitans are equally efficient antigens in enzyme-linked immunosorbent assay analysis J Clin Microbiol 200240 46405 DyeBAHerrera-AbreuMLerche-SehmJVlachojannisCPikdokenLPretzlB Serum antibodies to periodontal bacteria as diagnostic markers of periodontitis J Periodontol 200980 63447 NaitoYOkudaKTakazoeI. Immunoglobulin G response to subgingival gram-negative bacteria in human subjects Infect Immun 198445 4751 CraigRGBoylanRYipJMijaresDImamMSocranskySS Serum IgG antibody response to periodontal pathogens in minority populations: relationship to periodontal disease status and progression J Periodontal Res 200237 13246 van WinkelhoffAJLoosBGvan der ReijdenWAvan der VeldenU. Porphyromonas gingivalis, Bacteroides forsythus and other putative periodontal pathogens in subjects with and without periodontal destruction J Clin Periodontol 200229 10238 HaffajeeADSocranskySS. Microbial etiological agents of destructive periodontal diseases Periodontol 2000 19945 78111 HägewaldSBernimoulinJPKöttgenEKageA. Salivary IgA subclasses and bacteria-reactive IgA in patients with aggressive periodontitis J Periodontal Res 200237 3339 HenskensYMvan den KeijbusPAVeermanECVan der WeijdenGATimmermanMFSnoekCM Protein composition of whole and parotid saliva in healthy and periodontitis subjects. Determination of cystatins, albumin, amylase and IgA J Periodontal Res 199631 5765 ApatzidouDARiggioMPKinaneDF. Impact of smoking on the clinical, microbiological and immunological parameters of adult patients with periodontitis J Clin Periodontol 200532 97383