The incidence of long-term oral complications after hematopoietic SCT (HSCT) varies between 60 and 100%. The aim of this study was to compare the salivary secretion rate and the contribution of known risk factors for a low salivary secretion rate 1 year after HSCT in children conditioned with fractionated TBI (fTBI) and in children conditioned with single-dose TBI (sTBI). The study involved 44 patients, 27 conditioned with sTBI and 17 conditioned with fTBI. The unstimulated and stimulated salivary secretion rates (USSRs and SSSRs) were estimated before HSCT and at 1-year follow-up. Risk factors that may have influenced the salivary secretion rate were recorded. An SSSR of ⩽0.5 mL/min and a USSR of ⩽0.1 mL/min were chosen as cut-off points for salivary dysfunction. The median reduction in stimulated salivary flow 1 year after HSCT was 56% in the sTBI group and 12% in the fTBI group (P=0.003). The median reduction in unstimulated salivary flow 1 year after HSCT was 74% in the sTBI group and 33% in the fTBI group (P=0.003). In the multivariate model, a significant correlation between both sTBI (odds ratio (OR)=6.49, 95% confidence interval (CI)=1.40–30, P=0.014) and seropositivity of the recipient for 3–4 herpesviruses (OR=6.57, 95% CI=1.26–34, P=0.021) and a low stimulated salivary secretion rate (<0.5 mL/min) was found 1 year after HSCT.

33 patients undergoing colorectal surgery were treated prophylactically with aztreonam (19 patients), moxalactam (10 patients) and ampicillin plus sulbactam (4 patients). The effect on the normal oropharyngeal microflora and new colonization of oropharynx were studied during and after the treatment. Aztreonam had minor effect on the normal flora but induced colonization with staphylococci. Moxalactam suppressed gram-negative aerobic and anaerobic bacteria and promoted colonization with Candida albicans. Ampicillin plus sulbactam had marked effect on the normal flora and suppressed aerobic as well as anaerobic bacteria. New colonization with enteric gram-negative rods or fungi was observed in all patients.

Excessive blood loss is a serious but rare complication of orthognathic surgery. The aim of this study was to find out whether template bleeding time (TBT) could detect primary bleeding disorders in patients having orthognathic operations and how many false positive tests there were. We also examined the correlation between the perioperative bleeding rate and the result of preoperative TBT. Patients who had orthognathic operations at Karolinska University Hospital, Huddinge, from August 2001 to December 2006, were screened preoperatively by measuring TBT and the records of their operations were reviewed retrospectively. Prolonged TBT was recorded in 20 patients (13%), 19 female and 1 male (P=0.02). After further examination, 10 of these patients were found to have a primary disorder of haemostasis. No detectable reason for the prolonged TBT could be found in the others. There was no significant difference in perioperative bleeding rate between patients with prolonged and normal TBT or between TBT and perioperative bleeding rate. In conclusion, the routine use of preoperative TBT for haemostatic screening in orthognathic surgery cannot be recommended because of the large number of false positive results. There was no significant correlation between prolonged TBT and bleeding during orthognathic surgery.

The occurrence of oral penicillin-resistant viridans group streptococci (VGS) was studied in 50 patients with either newly diagnosed acute leukaemia or autologous peripheral stem cell transplants. One patient was excluded because of Staphylococcus aureus growth in the stem cell harvest. VGS were isolated from the oral cavity of 48 of the remaining 49 patients. Of these 48 patients, 12 (25%) yielded VGS resistant (MIC > 2 mg/L) to penicillin. These 12 patients had a higher frequency of septicaemia (p 0.04) and more days of treatment with trimethoprim-sulphamethoxazole (p 0.04) than patients who harboured susceptible or intermediately resistant VGS (MIC 2 mg/L). There were no other statistically significant differences between the two groups. It is important to be aware of the high level of penicillin resistance in oral VGS in patients with haematological disease, and this parameter should be considered when selecting antibiotic therapy for cases of septicaemia caused by VGS in immunocompromised patients.

Patients who were unable to have an implant of the maxilla without a bone graft were operated on by a modified sinus lift method, under local anaesthesia. Cortical bone grafts, harvested from the lateral side of the mandible, were forced horizontally into a slot, made in the lateral sinus wall after a mucosal sinus lift. The immediate stabilization of the graft permitted the dense packing of the space below the horizontal graft with bone chips, which probably contributed to the fast healing of the bone. We saw eight patients (10 sites) for long-term follow-up. Two or three Brånemark standard implants were inserted into the grafted area 3 months postoperatively. The abutments were connected 5 to 6 months after implantation. We have now adopted this modified method as the standard for bone grafting to the maxillary sinus in both of our maxillofacial centres.

The use of antibiotic prophylaxis in patients predisposed to infective endocarditis (IE) who are undergoing oral surgery procedures is widely accepted. Because there have been no controlled clinical trials of antibiotic regimens for the prevention of endocarditis in humans, recommendations are based on the results of prophylaxis studies in animal models of endocarditis, in vitro susceptibility data of pathogens that cause endocarditis, procedure-related studies of bacteremia, and studies of the efficacy of antimicrobial prophylaxis for prevention of postsurgical bacteremia.

The antimicrobial susceptibilities of 311 strains of anaerobic bacteria and 140 strains of aerobic bacteria, isolated from blood after dental extraction, were determined by the E test and compared with the results obtained by the agar dilution method on PDM-ASM 2 agar. E test MICs agreed within +/-1 dilution step to the agar dilution MICs in 93%, 52%, 90%, 94% of the tested anaerobes for penicillin V, cefaclor, clindamycin and erythromycin, respectively. For aerobic bacteria the agreement was > or = 90% for the antibiotics tested. The in vitro activities of penicillin V, clindamycin and erythromycin were higher than the activity of cefaclor against the majority of bacteria tested.

It is now generally accepted that adherence of microorganisms to various components of cardiac valve surfaces or vegetation lodging on the heart valves is an important early event in the pathogenesis of infective endocarditis. 120 clinical isolates of S. aureus obtained from patients with endocarditis and wound infections and from nasopharyngeal carriers were quantitatively analysed in vitro for their ability to bind to fibronectin and to produce protein A and alpha-toxin. Both cell-bound and extracellular protein A were measured and alpha-toxin was determined as antigen and as haemolytic activity. The highest fibronectin binding ability was found in carrier strains while no significant differences between strains were observed regarding the production of protein A. Strains isolated from patients with endocarditis produced significantly lower amounts of alpha-toxin than did strains from the other two groups. An inverse relationship between the production of protein A and of alpha-toxin was noticed in the material. Animal passage of five strains in an experimental endocarditis model showed a good reproducibility of the test systems and one strain was upregulated in its fibronectin binding ability and in alpha-toxin production. These in vitro results indicate that the fibronectin binding ability is not the decisive adherence factor and question the role of alpha-toxin as a virulence factor in endocarditis.

The adherence characteristics in vivo and virulence of two isogenic strains of Staphylococcus aureus differing in fibronectin binding were compared in a rat model of catheter-induced infective endocarditis. No differences were found between the two strains. The results strongly point to the multifactorial nature of bacterial adherence to damaged heart valves and suggest that other binding functions can compensate for the lack of fibronectin binding in S. aureus.

The role of Staphylococcus aureus collagen binding in the development of experimental endocarditis was studied. Two isogenic strains of S. aureus, 1 carrying an insertional inactivation of the gene encoding collagen-binding protein, were compared in a rat model of catheter-induced infective endocarditis (i.e.). Separate groups of rats with traumatized aortic valves were intravenously challenged with 1 of the strains. In rats sacrificed 24 h after inoculation, the collagen-binding strain significantly outnumbered the mutant strain (P < .001); however, 1 h after challenge, there was no difference in numbers of the 2 strains. The results were substantiated, using a 1:1 mixture of the parent strain and the mutant as an inoculate. Our findings suggest that collagen binding of S. aureus is important in the sustenance of experimental IE and plays a limited role during the initial attachment of the microorganism to traumatized aortic valves.

This study was undertaken to investigate the effects of prophylactic administration of cefaclor on bacteremia after dental extraction. Thirty-nine patients were randomly assigned to receive either 1 g cefaclor (19 patients) or placebo (20 patients) 1 h prior to dental extraction. Blood samples for microbiological investigation were collected before, during, and 10 min after surgery, and were processed by lysis filtration under anaerobic conditions. The incidence of bacteremia with viridans streptococci was 79% in the cefaclor group and 50% in the placebo group during extraction. No difference in the incidence or magnitude of bacteremia was observed when the two patient groups were compared.

A mutant deficient for the 19-kDa extracellular fibrinogen-binding protein (Fib) from Staphylococcus aureus has been constructed. The gene was inactivated by allele replacement. A 2.0-kb fragment from transposon Tn4001 carrying the gene for gentamicin resistance was inserted into the gene encoding Fib (fib). The genotype was verified by PCR analysis, and the loss of Fib was demonstrated by Western blotting (immunoblotting). The mutation has not altered the ability of the strain to bind to fibrinogen or fibronectin compared with that of the isogenic parental strain, FDA486. The mutant, designated K4.3, was compared with strain FDA486 in a wound infection model in rats. Sixty-eight percent of the rats challenged with parental strain FDA486 developed severe clinical signs of wound infection, whereas only 29% of the animals challenged with isogenic mutant K4.3 showed severe symptoms (P < 0.01). The weight loss of animals infected with the wild type was also significantly different from that of animals infected with the mutant strain. The result demonstrates that the extracellular 19-kDa fibrinogen-binding protein from S. aureus contributes to the virulence in wound infection and delays the healing process.

Streptococcus mutans P1 antigen functions as an adhesion factor for binding to salivary pellicle on tooth surfaces. It induces increased antibody titres in patients with Strep. mutans endocarditis. A mutant of Strep. mutans deficient in the function of the gene (spa P) encoding the surface antigen P1, and its isogenic parental strain, were used in a rat endocarditis experiment. Absence of P1 did not decrease adhesion to vegetations determined 1 h after intravenous infection. The number of bacteria recovered from valvular vegetations after 48 h from animals with manifest endocarditis did not differ between the strains. Consequently, the PI antigen appears to be unimportant both for adhesion and virulence in endocarditis caused by Strep. mutans.

Hematogenous osteomyelitis was produced in the tibia or the mandible of rats by drilling a hole into the bone, injecting sodium morrhuate, and inoculating Staphylococcus aureus Phillips into the femoral vein. Animals were sacrificed after 2 weeks and examined. The infection was characterized grossly and radiographically by bone deformation, histopathologically by a characteristic suppurative reaction, and microbiologically by the recovery of S. aureus Phillips from the infected tissue. These findings indicate that the model mimics human osteomyelitis with respect to its inflammatory bone changes. In contrast to earlier rat models in which bacteria were injected directly into the bone, this new experimental model allows study of the initiating events of osteomyelitis such as bacterial attachment and might assist as a model for both prophylactic and therapeutic trials.

Patients with mandibular osteomyelitis had quantification of 10 antibodies against certain bacterial proteins and polysaccharides. Sera from 31 patients with acute or chronic osteomyelitis of the mandible and from 17 healthy controls were analyzed. Some patients showed low levels of investigated antibodies in general and a lack of specific antiteichoic acid antibodies, as well as of different antipneumococcal antibodies particularly. Two patients with therapy-resistant osteomyelitis showed IgG(2) and IgG(3) subclass deficiency. They had replacement therapy with intravenous 10 or 15 gm immunoglobulin every 3 weeks for 6 months. Both patients showed considerable improvement in their clinical symptoms after treatment with immunoglobulin. This study indicates that impaired humoral immune response may be of importance in subgroups of patients with osteomyelitis of the mandible.

A total of 67 teeth in 64 patients were treated with apicectomy and retrograde fillings. They were randomized to receive fillings of amalgam or glass ionomer cement in a comparative clinical study. Healing was evaluated clinically and radiographically after 1 and 5 years. Evaluation showed no difference in healing capacity between the two materials. Overall success rates in both groups were registered as 90% at 1 year and 85% at 5 years. Contamination with blood or saliva during insertion of the filling material did not affect healing adversely. The study shows that the 5-year follow-up result can be predicted in more than 95% of the cases at the 1-year follow-up. It can be concluded that glass ionomer cement is a valid alternative to amalgam as an apical sealant after apicectomy with equally good long-term clinical results.

This study evaluated whether cytomegalovirus (CMV) neutralizing capacity affected shedding of CMV in saliva in human immunodeficiency virus type 1 (HIV-1)-infected patients and mapped specific epitope reactivity of CMV IgG antibodies. Total CMV IgG titers were significantly higher in symptomatic than in asymptomatic HIV-1-infected patients or controls. All CMV-seropositive patients had neutralizing antibodies to CMV. Shedding of CMV in the saliva of AIDS patients occurs despite the presence of serum antibodies with a high capacity to neutralize autologous CMV isolates. The highest IgG reactivity against a CMV envelope protein (gp116), represented by a peptide, was found in patients with advanced HIV disease. In contrast, identical IgG reactivities against a peptide representing the CMV matrix protein were observed in healthy controls and HIV-1-infected persons.

The aim of this study was to evaluate whether HIV-1 or cytomegalovirus (CMV) may contribute to oral lesions frequently found in patients with the acquired immunodeficiency syndrome (AIDS). Saliva samples from 63 HIV-1 positive patients and 21 healthy controls were tested for the presence of HIV-1 and CMV using the polymerase chain reaction (PCR) and virus isolation. CMV IgG titres in serum were also compared in the different groups. HIV-1 RNA, but not DNA, was detected in saliva from 15% (9 out of 59) of the HIV-infected patients. There was no correlation between the presence of HIV-1 RNA and oral symptoms in the patients. CMV DNA was detected significantly more frequently in samples from HIV-1 seropositive than from seronegative patients. CMV was detected in saliva from AIDS patients more often than in saliva from patients with mild or no symptoms. CMV IgG titres were also significantly higher in symptomatic than in asymptomatic patients. There was a significant correlation between the presence of CMV DNA and necrotizing gingivitis, and oral Kaposi's sarcoma in the patients, and also between these lesions and the stage of disease. This does not prove that CMV causes these oral lesions, but a direct or indirect role for CMV cannot be excluded.

Sixty healthy patients were randomized to receive placebo, penicillin V (2 g), or amoxicillin (3 g) 1 hour before dental extraction was performed. Blood samples for microbiological investigation were collected before, during, and 10 minutes after surgery and were processed by lysis filtration under anaerobic conditions. There was no statistical difference among patients in the placebo group, the penicillin-V group, and the amoxicillin group in terms of incidence or magnitude of bacteremia due to viridans streptococci or anaerobic bacteria during extraction or 10 minutes after the procedure. The overall incidence rates of bacteremia after dental extraction were 95%, 90%, and 85%, respectively, for the three groups. For > 90% of 126 strains of viridans streptococci tested, the MICs of penicillin V and ampicillin were < or = 0.125 mg/L. The protective effect of prophylactically administered penicillins must be due to interference with crucial steps in the development of endocarditis (other than the transient bacteremia that occurs initially).