Saliva plays a major part in the health of the mouth and any changes in its amount or quality may alter the oral health status. There have been relatively few studies on the oral health of asthmatic patients. Their findings, indicating an increased risk of oral diseases in asthmatic patients, were mainly obtained from children and adolescents. Young asthmatic patients suffer more from dental caries and periodontal diseases than non-asthmatic individuals. Also there are reportedly differences between asthmatic and non-asthmatic individuals in salivary flow rate and composition. In summary, there are no studies on salivary variables or oral health in adult asthmatics. Thus, our aim now was to compare the composition of stimulated whole saliva from adult asthmatic patients and non-asthmatic individuals.

Twenty-six asthmatics were enrolled from among outpatients visiting the Clinic of Pulmonary Diseases, Tampere University Hospital. The inclusion criteria were age ranging from 25 to 50 years and diagnosed asthma. Patients with medication for psychiatric diseases were excluded as well as those with diabetes or any other disease that may directly or indirectly affect the mouth. As a control, 33 non-asthmatic individuals were enrolled from either the Occupational Health Care Centre of Tampere City (n=27) from among those who came for regular check-ups, or from medical students at Tampere University (n=6). The inclusion criteria were the same as for the asthmatics except the diagnosed asthma. Also the exclusion criteria were the same.

In the asthma group, 21 participants regularly used inhaled steroids. Twenty-five also inhaled short-acting ?₂-agonists ‘as needed’. In addition, the asthmatics used the following medications: antibiotics (n=1), ?-blocker and tolfenamic acid (n=1), muscle relaxant (n=1) and sucralfate mixture (n=1). In the control group, eight participants used the following medications: oral contraceptives (n=5), antifungal treatment (n=1), antihypersensitive and lipid-lowering drugs (n=1) and, irregularly, diuretics and analgesics (n=1). There were no major differences in smoking habits between the two groups and the frequency of reported alcohol intake was almost the same. The study was carried out according to the Declaration of Helsinki and was approved by the ethics committee of Tampere University Hospital. Oral examinations were carried out according to the WHO guidelines. In the examinations the DMFT and periodontal status were recorded. For periodontal status a scoring system similar to the CPITN index was used. Every tooth was given a periodontal score ranging from 0 to 4: 0=no bleeding after probing, calculus or periodontal pocket found (healthy gingiva); 1=gingival bleeding after probing (mild periodontal inflammation, gingivitis); 2=presence of dental calculus (moderate periodontal inflammation); 3=periodontal pocket depth 4–6 mm (severe inflammation), and 4=periodontal pocket depth > 6 mm (very severe inflammation). The individual periodontal status was expressed as a PSI calculated as follows. In each participant the frequencies of scores were calculated. Then the numbers of teeth with inflamed periodontal tissues (score 1) were summed and the sum divided by the number of all remaining teeth. The quotient was expressed as a percentage. The PSI score was developed for this study.

Paraffin-stimulated whole saliva was collected into chilled, graduated glass tubes at 8.00–10.00 a.m. Saliva was collected during 5 min from all the participants and the collected volumes were measured. The participants were not allowed to use any drugs and refrained from smoking, eating and drinking for 1 hr before saliva collection. Immediately after the collection, 100 ?l of uncentrifuged saliva was transferred to a plastic tube containing fresh tryptic soy broth (Oxoid, Basingstoke, U.K.) supplemented with 20% glycerol (used for microbiological analysis). The samples were stored frozen at ?20°C for 1 month before bacterial cultivation. Another portion (50 ?l) of uncentrifuged saliva was separated and transferred to Eppendorf vials (Plastic Trade, Helsinki, Finland) for the analysis of Ca²⁺ concentrations. The lysozyme activity, the concentrations of total protein, lactoferrin, myeloperoxidase and salivary peroxidase, as well as SCN^?, Na⁺ and K⁺ were determined after centrifugation at 12,000×g (Sorvall Superspeed RC-2B centrifuge) for 10 min at 4°C. Portions of uncentrifuged or centrifuged saliva needed for the analysis were divided into separate Eppendorf vials and stored at ?20°C until analysed.

The total protein concentration was measured by the method of with bovine serum albumin (Sigma Chemical Co, St. Louis, MO) as a standard. Salivary peroxidase and myeloperoxidase concentrations were quantitated with immunometric assays using biotinylated antibodies and avidin–alkaline phosphatase label (Cappel, Organon Teknika Corp., West Chester, PA) for the detection . The standards, purified human leucocyte myeloperoxidase and bovine milk lactoperoxidase (Sigma), were used as immunogens in raising antibodies . Bovine milk lactoperoxidase and human salivary peroxidase are immunologically cross-reactive . The lactoferrin concentrations were determined by an immunometric assay using a biotinylated antibody and avidin–biotin–peroxidase complex (Vector Laboratories, Burlingame, CA) . Human colostral lactoferrin (Sigma), further purified by affinity chromatography, was used both as an immunogen in raising the antibody and as a standard in the assay. The immunometric assay for lysozyme was similar to the lactoferrin assay. Standards (human urine lysozyme, a kind gift from Dr Pertti Mörsky, Tampere, Finalnd) and samples were first incubated on microplates (Maxisorb, NUNC, Roskilde, Denmark) precoated with antihuman lysozyme (Dakopatts, Glostrup, Denmark). Then biotinylated antilysozyme, avidin–biotin–peroxidase complex and enzyme substrate were sequentially incubated on the plate. The absorbances in the immunometric assays were measured with a microplate spectrophotometer (Multiskan, Eflab Oy, Helsinki, Finland). The concentrations of the electrolytes, Na⁺, K⁺ and Ca²⁺, were assessed by atomic absorption (Atomic Absorption Spectrophotometer 460; Perkin Elmer, Norwalk, CT). SCN^? ions were quantified by the ferric nitrate method .

Before the cultivation of salivary microbes the tubes with tryptic soy broth were thawed and vortexed thoroughly for 1 min. After serial-fold dilutions, the bacteria were plated as follows. Mutans streptococci on mitis salivarius bacitracin (Difco Laboratories, Detroit, MI) agar plates; these plates were supplemented with 20% sucrose (BDH Chemicals Ltd., Poole, U.K.) and 0.5 ?g/ml bacitracin (Sigma). The plates were incubated for 3 days in a 7% CO₂ atmosphere at 37°C. Lactobacilli were cultivated on Rogosa SL agar plates (Difco) and incubated anaerobically (80% N₂, 10% H₂, 10% CO₂ atmosphere) for 3 days at 37°C. The total anaerobic flora was determined by plating samples on blood agar plates containing 5% sheep blood and incubating anaerobically (80% N₂, 10% H₂, 10% CO₂ atmosphere) for 2 days at 37°C. The number of Candida spp. in whole saliva was determined by plating samples on Sabouraud agar plates (Bacto®, Difco). The plates were incubated aerobically for 24 hr at 37°C. After appropriate incubation times the number of c.f.u./ ml saliva was determined.

The results from the analysis of saliva samples are expressed as crude concentrations. The mean values and SD were calculated for each variable measured by at least an interval scale. The statistical significance of the differences between groups was tested with Student’s t-test for unpaired samples. Moreover, the 95% CI for the differences between the means was calculated. All computations except 95% CI were made with the BMDP statistical package (Version 1990). The 95% CI were calculated with a Confidence Interval Analysis program running in a PC. In the statistical tests a two-sided p-value of 0.05 was considered statistically significant.