Identification of risk patients in oral implantology (II)

Long term studies in both fully edentulous and partially edentulous patients have emphasised the importance of following the principles of osseointegration for predictably successful implant treatments.

However, other variables should be considered and the aim of this review is to evaluate not only the influence of general health on the success of implants, but also the effect of implants on general health. Scientific evidence is sometimes limited and this paper is restricted to consideration of risk factors additional to those developed in this issue by ^{Sanz and Etienne (1998)}.

The influence of systemic factors on implant therapy

^{Maeglin (1983)} is quoted by ^{Schroeder (1991)} as referring to the systemic contra-indications for the placement of implants. They will be short-lived and cause intermittent or chronic infections if they are considered in cases of systemic bone disease, diseases of the endocrine and haemopoetic systems, rheumatic, cardiac, kidney and liver diseases, allergic conditions, the immuno-compromised and where focal infection is suspected. However, ^{Feigel (1985)} believed that patients could be considered for implants if their systemic disorder is under control and monitored. ^{Smith et al. (1992)} were also opposed to the concept of absolute contra-indications. In their study of 313 Brånemark implants in 104 patients with systemic disease, 90 % of the subjects had an altered periodontal state. Nevertheless, there was failure of osseointegration in only 13.5 % of the individuals.

^{Misch (1993)} also cited factors which have been implicated in failure to heal and we will restrict our consideration to : age, malnutrition, vitamin and zinc deficiencies, anaemia, kidney and liver disease, malignant disease, steroid therapy and chemotherapy.

Absolute surgical contra-indications

^{Barco (1991)} considered that despite a 3 %-19 % prevalence of endocarditis following dental treatment, 32 % of cases not caused by streptococci or staphylococci were associated with dental treatment. The general opinion amongst practitioners presented with patients with heart disease is guarded and that because of serious medico-legal consequences, implants are generally contra-indicated. The recommendations of the 5th consensus conference on anti-microbial therapy (1992), supports this view and state that implants and periodontal surgery are not advised in patients at high risk of infectious endocarditis.

However, we can debate the rationale for therapeutic decisions in the case of partially edentulous cases. The peri-implant soft tissues, being histologically comparable to scar tissue, are poorly vascularised and their potential for defence is less favourable than tissues surrounding teeth. Peri-implant infections probably originate from the presence of gingivitis or periodontitis around the natural teeth ^{(Gouvoussis et al., 1997)}. Nevertheless, it has not been demonstrated that peri-implant destruction progresses any quicker than periodontitis. These authors also state that Actinobacillus actinomycetemcomitans (Aa) and Eikenella corrodens (Ec) can be recovered from all implant sites if they have been isolated from periodontal pockets, Prevotella intermedia (Pi) and Fusobacterium nucleatum (Fn) from 83 % and Porphyromonas gingivalis (Pg) from 75 %. Therefore, the colonisation of implants is possible, but the implant surface has its own characteristics and ^{Edgerton et al. (1996)} have demonstrated that the adhesion to titanium of normally weakly adhesive streptococci (S. anginosus, S. oralis, S. salivarius) can be enhanced by the salivary pellicle. These authors consider that the differences in colonisation, when compared with natural teeth, could thus be explained and that ideas of peri-implant health or disease do not necessarily lie in the quality of the periodontal flora.

Bearing in mind the risk of bacteraemia during dental treatment, it is probable that patients with valvular disease also carry putative periodontal pathogens. This means that if one accepts the same criteria as for implants, patients with poor oral hygiene should have all teeth removed. ^{Misch (1993)} has a subtly differing opinion : « In some patients, implant therapy could be contra-indicated because of an increased risk of endocarditis ». He states that it is possible to experience a transitory bacteraemia in an edentulous patient during mastication, brushing or with peri-implantitis. In order to reduce these risks, these patients must have a satisfactory level of oral hygiene, and an adequate width of attached gingiva.

^{Wahl (1994)} emphasised that the American Association of Cardiology certainly recommends antibiotic cover before dental treatment, but without the evidence of any controlled studies. The author underlines that ethical problems associated with such a study militates against those responsible, because it is debatable whether antibiotics should be prescribed without evidence of the risks and cost involved. Similarly, ^{Bor and Himmelstein (1984)} consider that antibiotic cover (penicillin) for patients with a prolapsed mitral valve carries a 3 times greater chance of death because of anaphylaxis than of developing endocarditis in the absence of antibiotic cover.

^{Smith et al. (1992)} do not indicate any particular problems after placing implants in patients with heart disease and ^{Wahl (1994)} stated that the degree of risk varies according to nature of the heart disease. The lack of studies on this subject is, without doubt, due to the major risk posed by infective endocarditis, in comparison with a non-essential dental treatment.

We can also pose the question of those patients with implants who develop heart disease later and who may become high risk patients if, for example, they are fitted with a valvular prosthesis. ^{Zackin (1997)} cites the example of a serious case of infectious endocarditis which followed a maintenance periodontal treatment despite adequate antibiotic cover, as recommended by the American Association of Cardiology. In fact, this patient was required to have intravenous antibiotic, given in hospital immediately prior to the prophylaxis and eight hours later. The author underlines ^{Barco's (1991)} observations in connection with the prevention of endocarditis and the occasional ineffectiveness of systemic antibiotics, even when the micro-organisms are sensitive to them. The consequences can be very serious and ^{Cantrell and Yoshikawa (1980)} observed an increase in fatalities from infective endocarditis from 40 % to 70 % in patients over the age of 60 years. ^{Sixou et al. (1993)} emphasised the evidence provided in the literature and the potential of bacteria in the sub-gingival flora such as Aa, Ec, Cardiobacterium hominis and Capnocytophaga to cause infectious endocarditis, if the periodontal risk cannot be dealt with.

Syndromes affecting the residual bone and osseointegration

The idea of bone quality is not very satisfactory for the scientist but is important for the clinician (^{fig. 1} and ²). ^{Sennerby et al. (1992)} suggested the hypothesis that differences of bone quality between cortical and medullary bone can be detected at ultra-structural level. Therefore, these authors consider that resistance to extraction forces could be due to the thickness and biomechanical characteristics of the amorphous layer as well as amount of compact bone in contact with the implant.

Osteoporosis is defined as a generalised reduction in the mineral content of bone which can lead to fractures without any other anomaly in the chemistry of the bone ^{(Harrison, 1989} ; ^{Resnick, 1988)}. The condition occurs most frequently in women because of the reduction in oestrogen after menopause. The asymptomatic presence of osteoporosis is from 25 % in the age group 45-54 years to 39.2 % between 55 and 64 years ^{(Mangaroo et al., 1985)}. In women of 80 years, the loss of trabecular bone may thus reach 40 % compared with 27 % for a man of the same age.

Age changes are also associated with osteoporosis and ^{Roberts et al. (1992)} pointed out the increased risk factors :

1) Reduction in vitamin D synthesis ;

2) Inhibition of absorption of Ca⁺⁺ ;

3) levated levels of parathormone ;

4) A low level of calcitonin ;

5) Increased bone turnover ;

6) Pharmacologically active drugs (glucocorticoids, anticonvulsants, methotrexate, cyclosporin, lithium, tetracycline, aluminium based antacids, nicotine, heparin).

^{Gruber et al. (1996)} underlined the difficulties encountered in implantology with senile osteoporosis which are central to and due to insufficient apposition of bone during remodelling. The bony trabeculations are thinner and their resistance to biomechanical forces reduced when compared with younger subjects.

This loss of bone trabeculation can be detected at an early stage in the course of a pre-implant radiographic survey. The process of osseointegration which depends, in part, on the ability of the host bone to heal, on its quality and quantity could, in theory, be altered ^{(Dao et al., 1993)}. In a study undertaken on rabbit tibia, ^{Mori et al. (1997)} demonstrated that osteoporotic bone could affect the healing time after insertion of implants as well as the degree of osseointegration obtained.

In a study published in ^{1993, Dao et al}. evaluated the risk according to age (^{fig. 3} and ⁴) and sex in 45 women and 18 men over the age of 50, and 48 women and 18 men below the age of 50. The failure rates for men and women over 50 years of age were identical at 22.2 %, whereas the rates for women and men under the age of 50 were 18.8 % and 11.1 % respectively. According to this data, there was no significant association with age and the placement of implants in osteoporotic women does not seem to be a risk factor.

Many scientific studies have demonstrated that the healing capacity of osteoporotic bone is not reduced and that the whole of the bony skeleton is not affected in the same way. Moreover, if the reduction of bone mass can be determined by an orthopaedist at a given site, it does not necessarily mean that other sites will be similarly affected. The condition of bone at an implant site cannot be extrapolated from the condition at a distant site and the risk of failure is assessed mainly by radiography or by bone density during drilling. Nevertheless, ^{Kribbs et al. (1983)}, ⁽¹⁹⁸⁹⁾ showed a statistically significant correlation between the total reduction in bone mass and the density of mandibular bone.

According to ^{Jaffin and Bergman (1991)}, the failure rate is correlated not only with the quality of bone, but also with the position of the implant in the arch. They state that the failure rates for type IV bone are 44 % in the maxilla, 37 % in the posterior part of the mandible (^{fig. 5}, ⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹², ¹³ and ¹⁴) and 10 % in the anterior part. For the other types of bone (types I, II and III) the failure rates were 3.6 % in the maxilla, 6.8 % in the distal part of the mandible and 1.2 % in the anterior part.

^{Friberg et al. (1991)} found an increase in failure rate in patients with poor bone quality and an insufficient volume of bone. In a report of cases published in 1994, this author also concluded that osteoporosis need not be a contra-indication for implant treatment. According to ^{Jemt et al. (1992)}, treatment is possible and the short term prognosis is favourable as determined in a group of 92 patients presenting with severe maxillary bone resorption. Nevertheless, the failure rate was 18.9 % after 5 years in an area of bone loss supporting an overdenture, against a rate of only 3 % for conventional prosthetic restorations ^{(Jemt and Lekholm, 1995)}.

Implant success in patients suffering from osteoporosis necessitates a longer healing period, hyperbaric oxygen therapy and monitoring of the osteoporosis itself ^{(Fujimoto et al., 1996)}. Long term studies demonstrate that satisfactory implant treatment with a favourable outcome can be undertaken in older patients who have a strong probability of osteoporosis ^{(Jemt, 1993} ; ^{Adell et al., 1990)}. These data contrast with those in orthopaedics where the failure rates of orthopaedic implants for the treatment of osteoporotic fractures is high and increases with time ^{(Noble, 1983)}.

The concentration of extracellular Ca is regulated not only by parathyroid hormone, but also by vitamin D, prostaglandins, lymphocytes (by release of osteoclast activating factors), insulin, glucocorticoids and oestrogens.

Hyperparathyroidism is caused by overactivity of the parathyroid glands. Severe forms of this disease cause kidney, intestinal and bone disorders. Alveolar bone can be affected before the other bones of the skeleton, leading to total loss of the teeth. Radiographic examination reveals loss of the lamina dura and bone trabeculation resembling ground glass. Milder forms are usually asymptomatic and are not absolute contra-indications for the placement of implants but the surgical treatment must be undertaken under the supervision of a properly qualified team. These patients must be regularly monitored, not only in connection with their implant therapy, but also for their general medical management.

Radiotherapy

In recent years, many advances have been made in the treatment of head and neck tumours. These involve new techniques of three-dimensional radiotherapy combined with surgery. Survival rates for these patients have improved but the consequences of extensive maxillary resections for oro-pharyngeal cancer include a greater functional deficit and psychological problems. Because of anatomical problems in the mandible, a fragile mucosa and poor muscular function, it is frequently not possible to undertake conventional prosthetic treatment. In these situations, implants can be used to support prostheses but the risks to the irradiated tissues must be appreciated (^{fig. 15a}, ^15b, ^16a, ^16b, ¹⁷, ¹⁸, ^19a, ^19b, ^19c, ²⁰, ²¹, ²², ^23a and ^23b)..

^{Granström et al. (1992a)} described the early effects on the soft tissues and the later effects on bone. The early effects include xerostomia, dermatitis and mucositis, whereas the later effects are demineralisation, fibrous scar tissue, increased susceptibility to infection and ischaemic necrosis. The mandible is more susceptible to necrosis because of its compact bone and vascularisation by terminal arteries. To reduce these secondary effects, ^{Marx and Johnson (1987)} advocate the extraction of teeth and excision of tumours to be ideally at least 21 days prior to the radiotherapy.

^{Hansson et al. (1990)} found that the medium term evaluation of success for bone grafts or implants in irradiated subjects is difficult because of low survival rates. Also, the surgical procedure is demanding because it must cause as little trauma as possible to the bone in order not to reduce its healing potential. In the case of irradiation of the facial bones, the average failure rate after the first three years is usually quoted as 35 %. Implant loss being encountered in the following descending order : frontal bone, zygoma, mandible, maxilla and temporal bone ^{(Granström et al., 1993)}.

^{Arcuri et al. (1997)} emphasised not only the beneficial effects of radiotherapy in the treatment of cancers of the oral cavity and oro-pharynx, but also the deleterious effects on the healthy tissues. The treated sites are profoundly affected. Moreover, surgical interventions entail reduction in the depth of the vestibular sulcus as well as changes in the hard and/or soft tissues which limit the options available for conventional prosthetic treatment. According to ^{Marx and Johnson (1987)}, osteoradionecrosis, defined as death of bone due to radiotherapy, is only the final result after a series of attacks, first at a cellular, then at a tissue level, whilst the complications of osteoradionecrosis can lead directly to the death of the patient. The incidence of complications for radiation levels over 50 Gy will be, according to ^{Adamo (1979)} 81.1 %, whilst ^{Granström (1992b)} found no complications for doses below 48 Gy. Radiation has variable effects according to the tissue exposed to it, since ^{Ueda et al. (1993)} observed a marked inhibition of the healing of skin wounds with doses over 20 Gy, whereas ^{Kluth et al. (1988)} did not find any increase in osteoradionecrosis with increased radiation. In a study by ^{Asikainen et al. (1998)}, dogs had three weekly sessions of irradiation separated by one month, followed 2-3 months later by the placement of (non-buried) implants which were loaded after 4 months, for a period of 6 months. The authors showed the importance of divided doses in increasing the tolerance to bone remodelling and to the forces of mastication. The dose should be between 40-50 Gy because at 60 Gy division of the dose does not prevent the loss of implants.

The need to separate the bone surgery from the irradiation was considered by ^{Jacobsson and Albrektsson (1986)}. They placed cylindrical chambers in rabbit tibia, at the time of a single radiation dose of 15 Gy, 18 weeks later or 12 months later. The chambers were removed 4 weeks after their placement, which is sufficient time for complete new bone formation to occur in non-irradiated bone. The reduction in bone formation, evaluated by microdensitometry was 72 %, 57.1 % and 24.4 % respectively. The authors concluded that the ability of the bone to regenerate had not fully recovered after one year and that the results of implants or major bone reconstruction are more predictable when delayed in relationship to the irradiation. In contrast, ^{Marx and Johnson (1987)}, after biopsying irradiated bone, suggested placing implants 1-6 months after the irradiation because the reduction in vascularity (and perfusion of the tissues) and the increase in fibrous tissue increases the risk of osteoradionecrosis. These time constraints were not confirmed by ^{Jisander et al. (1997)} who found a cumulative success rate of 97 % in the mandible and 92 % in the maxilla despite implantation taking place 18-228 months after the irradiation.

Radiotherapy interferes with the quality of osseointegration because it took 54 % less force to unscrew an implant from rabbit tibia placed 8 weeks earlier, compared with the control ^{(Johnsson et al., 1993)}, whilst ^{Hansson et al. (1990)} did not notice any appreciable difference between immediate placement of the implant compared with a delay of 3 months.

This effect on healing was also illustrated by ^{Hum and Larsen (1990)} on rabbit bone irradiated two weeks after placement of an implant. After 8 weeks healing, they observed osseointegration over 76.2 % of the surface, compared with 94.8 % for the controls.

In humans, ^{Jacobsson et al. (1988)} had to remove 5 out of 35 implants following lack of osseointegration in the temporal bone or lower border of the orbit. ^{Parel and Tjellstrom (1991)} obtained data from different centres in the USA and Sweden, showing success rates of 61 % in 27 patients after the placement of 108 implants in irradiated facial bones. The failures were mainly in the orbital region. ^{Tolman and Taylor (1996)} found a survival rate of 85 % after 6-30 months for implants supporting auricle, orbital or nasal prostheses in bone irradiated mostly 12 months earlier. The less satisfactory results were obtained in orbital bone which is thinner and denser.

Taking into account the density of bone and the terminal arterial blood supply, the placement of implants in irradiated mandibles is supposed to give inconsistent results. ^{Taylor and Worthington}, from 1993 indicated successful results for a period of 3-7 years in Brånemark implants in 4 mandibles irradiated in conjunction with hyperbaric oxygen therapy (HBO). The authors recommend a careful surgical technique, under general anaesthesia rather than local, because the placement of implants with the latter involves the use of vasoconstrictors. ^{Watzinger et al. (1996)} noted a success rate of 87.8 % after 3 years in 60 mandibular IMZ implants placed at least 12 months after a similar single dose of radiotherapy. The results in a group treated with bone grafts were less favourable (58.3 %).

The need for HBO treatment has generated some controversy. HBO induces proliferation of capillaries and fibroblasts, as well as collagen synthesis in bone and the soft tissues, which reduces the risk of osteoradionecrosis ^{(Granström et al., 1992a)}. These authors reported a failure rate of 58 % in the maxilla and orbit, against 2.6 % where the bone had been irradiated and treated with HBO. ^{Ueda et al. (1993)} placed 21 implants in irradiated bone in 4 patients who had suffered cancer of the orbit, maxilla and mandible and only one implant failed (survival rate of 92.3 % at stage 2). ^{Johnson et al. (1994)} reported success rates greater than 80 % after 6 months loading following insertion into an irradiated and HBO treated mandible and 100 % in 6 patients followed up for 3-6 months after loading. The majority of 24 implants were placed in the mandible and 21 were coated hydroxyapatite ^{(Smatt et al., 1995)}.

The success rate can be assessed objectively by a histological study of the increase in osseointegration on irradiated and HBO treated bone and by improvements in the healing of soft tissues ^{(Larsen et al., 1993)}. After 8 weeks the unscrewing force required from rabbit tibia was increased by 44 % after hyperbaric treatment, compared with irradiated bone and by 22 % compared with the controls ^{(Johnsson et al., 1993)}.

^{Arcuri et al. (1997)} placed 18 implants in the irradiated and HBO treated mandibles of 4 patients. The radiotherapy took place between 1.25 and 11 years before the fitting of the implants and the HBO treatment consisted of 20 pre- and 10 post-operative sessions of 90 minutes each, at 2.4 atmospheres. Implants were only placed if bleeding could be elicited at the site. Osseointegration was favourable by the time of placement of the abutments (94 %) and these implants remained successful during the follow-up period ranging from 1 to 5 years (mean 37 months).

Further studies with a large number of cases are required in order to evaluate long term success rates. After placement of conventional prostheses, the implant option, after HBO treatments would reduce the risks of tissue trauma and osteoradionecrosis. However, in 20 patients who underwent radiotherapy without HBO, the survival rate at stage 2 was 100 % for auricular and nasal implants and 79 % for orbital implants. ^{Franzén et al. (1995)} placed 20 implants in only 5 patients who had radiotherapy to the mandibles for two years. This length of time was necessary to ensure remission of the carcinomas. Nineteen of the implants remained stable after 3-6 years observation. Despite successful treatment in two smokers, it is suggested that implant therapy must be contra-indicated in those patients who are smokers and have undergone radiotherapy. Moreover, they emphasise that there is a big difference between the prevention and treatment of osteoradionecrosis by HBO. In addition, osteoradionecrosis would be rare, mostly localised to the retromolar region which is relatively little used and one would not consider HBO treatment until after comparative studies have been undertaken and scientific evidence of the benefits obtained. ^{Larsen et al. (1992)} clearly showed that HBO treatment of rabbit tibia caused an improved contact at the bone-implant interface but there was no clinical or radiographic difference between the two groups.

These controversies could possibly apply to maxillary implants. ^{Ali et al. (1997)} treated 10 patients without HBO and followed up their prosthetic restorations for a mean period of 33 months. Thirty two mandibular implants were placed in 7 patients with a success rate of 100 %. In contrast, of 10 maxillary implants in 3 patients, 6 implants were lost (60 % failure rate) in 2 patients over a mean observation period of 31 months. The authors emphasised that the implants were short and there was definite overloading because obturators were fitted at the time of placement. They suggest placing the implants at the time of the primary ablative surgery, together with HBO to the maxilla, but no mention is made of increasing the number of implants in order to spread the occlusal load. ^{Jisander et al. (1997)} had a better success rate in 17 patients, whatever the location or the radiation dose, but the dose received was in excess of 50 Gy and the patients also received HBO. Finally, ^{Keller et al. (1997)} did not use HBO in the treatment of 19 tumour patients, but advocated a technique involving minimal trauma. Ninety eight implants were placed anterior to the mental foramen in the 19 patients, 16-168 months after the radiotherapy. After a mean period of 44 months in function, the survival rate was 99 % (one implant which had been placed in a class IV bone graft in the molar area was lost), 100 % for the prostheses and 89 % for bone grafts where radiotherapy had been undertaken before or after the grafting procedure. Although 4 patients developed osteoradionecrosis after radiotherapy, none suffered this complication after placing grafts or implants. The response of the irradiated soft tissues was acceptable, with or without keratinisation.

Two situations are also to be considered. With the increasing use of implants, it is likely that many of these patients over the age of 50 will receive radiotherapy for cancers of the head and neck. According to an in vitro model, the majority of « backscatter » radiation is absorbed in the first millimeter adjacent to the bone-implant interface and there is no increase in radiation at a distance of 3 mm ^{(Wang et al., 1996)}. According to ^{Granström et al. (1992a)}, if a patient requiring radiotherapy has implants in the area to be irradiated, the implants should be removed, but given the lack of scientific data, they suggest a pragmatic approach and only remove the prostheses, abutments and fixtures, leaving the implants in place. Indeed, removal one month prior to the radiation would be traumatic, this would leave the patient without a prosthesis and favour osteoradionecrosis. Complications occur mainly in the soft tissues (5 patients out of 11) and it is necessary to eliminate the irritant action of the abutments and connectors. HBO treatment, which reduces the effects of radiation on the soft tissues, should be considered ^{(Granström et al., 1993)}. The implant failure rate increased to 64.2 % in a group of 3 patients with osteoradionecrosis who had repeat radiotherapy from 2 months to 3 years for recurrent cancer of the face, one of whom received HBO treatment ^{(Granström and Tjellstrom, 1997)}.

The effects of radiation on immature bone have been assessed in rabbits by ^{Schon et al. (1996)}. Irradiation 5 days after placement in this animal would be equivalent to irradiation on the 21st postoperative day in man. If mature bone is relatively resistant, newly formed bone shows delayed osteogenesis, reduced bone formation and a reduction of the areas in contact with the implant. The bony changes are even more marked in the region of the apical threads of the implant. These effects could occur in the areas of bone synthesis near the screws and a bi-cortical anchorage in mature bone would be preferable.

Looking at these results, it is possible to say that radiotherapy is not a contraindication to implant therapy.

Systemic disorders and infection risk

Diabetes

Amongst the metabolic disorders, diabetes is one of the most common. Diabetics not only have a predisposition to infections but they also present with vascular complications. Protein synthesis is reduced, healing of soft and hard tissues is affected as well as regeneration of nerves and blood vessels.

Many epidemiological studies have shown that the prevalence of periodontal disease in diabetics is increased. ^{Westfelt et al. (1996)} demonstrated that diabetics, like non-diabetics, treated for moderate to advanced adult periodontitis, could maintain a state of periodontal health. In addition, the frequency of sites with signs of recurrent disease were similar in the two groups.

The periodontal diseases, like all other infectious diseases are capable of destabilising diabetes. The control of periodontal infection must, therefore, be an integral part of the general treatment of patients suffering from diabetes ^{(Grossi et al., 1997)}. The same approach can be adopted in implantology because there is no scientific evidence showing an increased risk of infection in diabetic patients by the fitting of implants. ^{Ellies (1992)}, in a retrospective study, observed that post-implant paraesthesia occurred in 4 out of 5 diabetic patients. He considered that the risk factors for paraesthesia in diabetics were related to a reduced resistance to infection, retarded healing, capillary angiopathy, neuropathy and arteriosclerosis.

Implant treatment in diabetics is not contra-indicated as long as the disease is under control. Antibiotics reduce the infection risk whilst relaxation and correct diet reduces the risk of hypoglycaemia. If the surgical procedure is long, intravenous insulin may be required, or half the normal dose is given on the morning of the surgery if there is a risk of oral administration being compromised. In these cases anti-inflammatory steroids are contra-indicated because of their effects on blood glucose levels.

If, at the initial examination and treatment planning stages, we take into account all the clinical parameters, we can predict a successful outcome and a good long term prognosis, in the same way that we can successfully treat periodontal disease in diabetic patients.

Immunodeficiencies

Several types of immunodeficiencies can be distinguished. Primary immunodeficiencies may be due to various alterations in the immune system, such as a reduction in the number of neutrophils as in cyclic neutropaenia, benign familial neutropaenia and other primary neutropaenias. Also, defects in neutrophil function are encountered in hyperimmunoglobulinaemia, chronic granulomatous diseases, kartegener syndrome, Chediak-Higashi syndrome, acatalasia and deficiencies in leucocyte adhesion. One can also cite Fanconi's anaemia, Down's syndrome and other multiple immunodeficiencies within the category of primary immunodeficiency.

In contrast, secondary immunodeficiencies are due to other causes such as malnutrition, psychological stress, pregnancy, diabetes or Crohn's disease. They can also be induced by vitamin C deficiency or smoking. It seems obvious that, depending on the severity of the immunodeficiency, the consequences that a peri-implant infection will have on the patient's systemic condition will have greater or lesser importance. There is no scientific evidence for any of the above to be absolute contra-indications for implant therapy. Nevertheless, it is admitted that one should do everything possible to reduce the possibility of infection, not only at the time of surgery, but also in the longer term.

HIV

It is interesting to note that significant psychological benefits can be obtained in sero-positive patients by the use of implants which enable patients to regain physical comfort. Progress which has been made in the treatment of these patients has increased their life expectancy. Implant treatment in sero-positive patients who are symptom-free provides long term benefits. Psychological factors must be considered and justify the fitting of implants, but the problems associated with the condition must not be ignored.

Cell counts of the numbers of CD4+ T-lymphocytes enable the progression of the immunodeficiency due to HIV to be monitored. The normal adult level is approximately 600 cells per c. mm and values below 500 per c. mm indicate immunosuppression. Opportunistic infections, such as oral candidosis, begin to appear with values below 400 per c. mm and there is the possibility of major infections with values below 200 per c. mm.

Surgical interventions would seem to be contra-indicated in patients with a CD4+ level below 500 per c. mm, but a large scale study did not demonstrate any major complications in patients whose CD4+ levels were below 200 per c. mm, when compared with a healthy population. Only 1 % experienced post-operative complications, such as excessive bleeding or poor healing ^{(Glick and Muzyka, 1993)}. Implants have been placed in HIV-infected patients with CD4+ levels above 200 per c. mm, haemoglobin above 10 mg/100 ml, a minimum platelet count of around 100,000 per c. mm and normal prothrombin levels, and they did not require longer periods for osseointegration, or experience greater amounts of bone loss ^{(Fielding et al., 1990)}.

However, ^{Ragni et al. (1995)} observed higher levels of post-operative infection in sero-positive haemophiliacs following orthopaedic surgery whose CD4 levels were equal to, or less than 200 per c. mm. Yet, ^{Unger et al. (1995)} did not observe any detectable clinical deterioration in the T4 levels of HIV patients and achieved satisfactory clinical results for arthroplasty of the knee, over a 9 year period.

However, these results must be treated with caution. Orthopaedic surgery involves sterile procedures (Altemeyer Class 1) which is definitely not the case in implant surgery where direct contamination with oral bacteria is possible. In addition, in cases of infection, there is persistent intimate contact between bacteria, the peri-implant tissues and even the implant itself. Alterations in host response have been described in HIV patients ^{(Ryder et al., 1988)}, notably in the level of polymorphonuclear cells ^{(Winckler and Hammerle, 1991)}, the mediators of inflammation such as interleukin 1Ê ^{(Lynch et al., 1992)} and TNF-α ^{(Jandinski et al., 1991)}. This diminished local response means that we must very carefully control all factors likely to lead to peri-implant infection.

Equally, care should be taken when prescribing any medication because of possible secondary effects or complications. Therefore, precautions should be taken when using antibiotics because of the risk of Candida albicans infections. They should be supplemented with anti-fungal preparations, or preferably, by using narrow spectrum antibiotics such as metronidazole which does not disturb the Gram + aerobic flora ^{(Winckler and Robertson, 1992)}.

Combined therapy with some analgesics (Di-antalvic^®, Dolosal^®) and some anti-inflammatory agents such as Feldène^® is contra-indicated with anti-proteases (Ritonavir^®).

Faced with HIV infection, it is the CD4 level which is the parameter to be considered in the pre-implant assessment. In view of the psychological benefits that can be obtained through the use of implants, they can be considered, in patients who request them, and whose CD4 level is above 400 per c. mm, without risk.

In patients who are at risk of infections, maintenance therapy plays, above all else, a fundamental role. Risk factors for activation of pathogenic bacteria and alterations to the plaque bacterial ecosystem must be evaluated and controlled.

The parameters to be observed are :

- the depth of the peri-implant « sulcus » which has been artificially produced in response to aesthetic requirements ;

- absence of pathogenic bacteria, especially in patients who are partially edentulous ;

- absence of supra-gingival plaque by the institution of oral hygiene measures, not only by the patient but also by professional personnel ;

- control of peri-implant inflammation, especially indicated by bleeding on probing.

Biocompatability

According to ^{van Steenberghe (1988)}, it is the passivity of titanium and the formation of an oxide layer which prevents its corrosion. This layer must not be scratched or affected by contact with chemically active ions or other metals. Titanium oxide is stable over a wide range of pH. Loss of titanium in biological fluids could be 10^-4 mm per year, but no systemic effect has been demonstrated.

The success of implants is characterised, in part, by the integrity of the soft tissue attachment with the implant abutment. Allergies to the metal of implants are possible and given the frequency of dermatitis due to contact with nickel, ^{Krauser (1986)} condemned implants containing alloys of nickel. The effect of corrosion of nickel and its release into the soft tissues produces an inflammatory reaction and may initiate an allergic reaction.

^{Smith et al. (1997)} considered the release of metallic ions following implantation and the fitting of overdentures. In their review, they emphasised that this release is influenced by the interaction of the tissues with the metal, by irregularities and porosities on the implant surface, the presence of debris and mobility of the implant. Blood analyses were undertaken up to 3 years after the placement of implants manufactured with a Ti-6Al-4V alloy, the surface of which was roughened to improve contact with bone. After 3 years, the level of Ti, Al and V was low and stable which indicates that there is no significant elevation of these elements in the circulating blood, but the authors emphasised that reliable information will only become available with the development of sophisticated analytical techniques.

Animal studies have shown no evidence of accumulation of these elements in organs ^{(Keller et al., 1985} ; ^{Lugowski et al., 1991)}, but ^{Smith et al. (1997)} emphasised that in animal studies, the implant surface area to weight ratio is 25 mm²/kg whereas in a human of 70 kg with 10 Brånemark-type implants, the ratio is 10 mm²/kg and in the case of a hip prosthesis, 290 mm²/kg. These differences, as with the type of prosthetic construction, could modify the release of ions.

As well as reactions to titanium which could lead to loss of osseointegration of implants, ^{Jacobs et al. (1991)} found twice as much titanium in patients with failing hip prostheses. Sensitivity to titanium could lead to implant failure ^{(Lalor et al., 1991)}.

Conclusions

Some systemic factors can contra-indicate the placement of implants because :

- the surgery itself represents a major risk from a systemic point of view ;

- systemic conditions may, to some extent, play an unfavourable role in osseointegration and soft tissue healing ;

- increased risk of infection represents an additional danger to the health of the patient.

If the practitioner chooses to place implants in these situations, he must take steps to minimise the risk of infections, as he does in periodontal treatment and modify the treatment in the light of these needs. It is necessary for him to avoid devices which are designed to improve aesthetics which lead to an increase in the depth of the peri-implant sulcus and false pocketing and which may create an environment favourable to the development of a pathogenic bacterial flora. Equally, he must ensure bacterial monitoring and sufficiently short maintenance intervals.

In these situations, certain contra-indications could be considered to be only relative. Nevertheless, before any therapeutic decision is made, it is important to ask ourselves whether we are able to prevent or to reduce the occurrence of infection in a compromised patient.

Demande de tirés à part

Daniel ETIENNE, Département de Parodontologie, Université Denis-Diderot, Paris VII, 5, rue Garancière, 75005 PARIS - FRANCE.

BIBLIOGRAPHIE

• ADAMO A, SZAL R. Timing, results and complications of mandibular reconstructive surgery : Report of 32 cases. J Oral Surgery 1979;37:755-763.
• ADELL R, LECKHOLM U, ERIKSSON B, BRANEMARK PI, JEMT T. A long term follow-up study of osseointegrated implants in the treatment of the totally edentulous jaw. Int J Oral Maxillofac Implants 1990;5:347-350.
• ALI A, PATTON D, El-SHARKAWI A, DAVIES J. Implant rehabilitation of irradiated jaws : a preliminary report. Int J Oral Maxillofac Implants 1997;12:523-526.
• ARCURI M, FRIDRICH K, FUNK G, TABOR M, LaVELLE W. Titanium osseointegrated implants combined with hyperbaric oxygen therapy in previously irradiated mandibles. J Prosthet Dent 1997;77:177-183.
• ASIKAINEN P, KLEMETTI E, KOTILAINEN R, VUILLEMIN T, SUTTER F, VOIPIO HM, KULLAA A. Osseointegration of dental implants in bone irradiated with 40, 50, or 60 Gy doses. An experimental study with beagle dogs. Clin Oral Impl Res 1998;9:20-25.
• BARCO C. Prevention of infective endocarditis : a review of medical and dental litterature. J Periodontol 1991;62:510-523.
• BOR D, HIMMELSTEIN D. Endocarditis prophylaxis for patients with mitral valve prolapse. A quantitative analysis. Amer J Med 1984;76:711-717.
• CANTRELL M, YOSHIKAWA T. Aging and infective endocarditis. J Amer Geriatr Soc 1980;31:216-222.
  Cinquième Conférence de consensus en thérapeutique anti-infectieuse de la Société de pathologie infectieuse de langue française. Paris, 27 mars 1992. Med Mal Infect 1992;22 (suppl. 3-12).
• DAO T, ANDERSON J, ZARB G. Is osteoporosis a risk factor for osseointegration of dental implants ? Int J Oral Maxillofac Implants 1993;8:137-144.
• EDGERTON M, LO S, SCANNAPIECO F. Experimental salivary pellicles formed on titanium surfaces mediate adhesion of streptococci. Int J Oral Maxillofac Implants 1996;11:443-449.
• ELLIES L. Altered sensation following mandibular implant surgery. J Prosthet Dent 1992;68:664-671.
• FEIGEL A. Die orale implantologie im « deutschsprachigen raum » : eine kritische betrachtung. Swiss Dent 1985;6:7-20.
• FIELDING AF, GLICK M, SALKIN M. Rational and criteria for dental implants in HIV-infected patients. UCLA symposium : Implants et patients édentés partiels, april 1990, Palm Spring, California, abstract 55.
• FRANZEN L, ROSENQUIST J, ROSENQUIST K, GUSTAFSSON I. Oral rehabilitation of patients with oral malignancies treated with radiotherapy and surgery without adjunctive hyperbaric oxygen. Int J Oral Maxillofac Implants 1995;10:183-187.
• FRIBERG B, JEMT T, LECKHOLM U. Early failures in 4 641 consecutively placed Brånemark dental implants : a study from stage 1 surgery to the connection of completed prosthesis. Int J Oral Maxillofac Implants 1991;6:142-146.
• FRIBERG B. Treatment with dental implants in patients with severe osteoporosis : A case report. Int J Periodont Rest Dent 1994;14:349-353.
• FUJIMOTO T, NIIMI A, NAKAI H, UEDA M. Osseointegrated implants in a patient with osteoporosis : A case report. Int J Oral Maxillofac Implants 1996;11:539-542.
• GLICK M, MUZYKA B. Evaluation of complications after dental treatment of severely immune suppressed HIV-infected patients. Presented at IX International Conference of AIDS, June 1993, Berlin, Germany. Abstract PO-B32-2239.
• GOUVOUSSIS J, SINDHUSAKE D, YEUNG S. Cross-infection from periodontitis sites to failing implant sites in the same mouth. Int J Oral Maxillofac Implants 1997;12:666-673.
• GRANSTRÖM G, JACOBSSON M, TJELLSTROM A. Titanium implants in irradiated tissue : benefits from hyperbaric oxygen. Int J Oral Maxillofac Implants 1992a;7:15-25.
• GRANSTRÖM G. The use of hyperbaric oxygen to prevent implant loss in the irradiated patient. In : Advanced osseointegration surgery : applications in the maxillofacial region. Worthington P, Brånemark P, eds. Chicago : Quintessence, 1992b;336-345.
• GRANSTRÖM G, TJELLSTROM A, ALBREKTSSON T. Postimplantation irradiation for head and neck cancer treatment. Int J Oral Maxillofac Implants 1993;8:495-501.
• GRANSTRÖM G, TJELLSTROM A, BRÅNEMARK P, FORNANDER J. Bone-anchored reconstruction of the irradiated head and neck cancer patient. Otolaryngol Head Neck Surg 1993;108:334-343.
• GRANSTRÖM G, TJELLSTROM A. Effect of irradiation on osseointegration before and after implant placement : A report of three cases. Int J Oral Maxillofac Implants 1997;12:547-551.
• GROSSI SG, SKREPCINSKI FB, DECARO Th, ROBERTSON DC, Ho AW, DUNFORD RG, GENCO RJ. Treatment of periodontal disease in diabetics reduces glycated hemoglobin. J Periodontol 1997;68:713-719.
• GRUBER H, SOLAR P, ULM C. Maxillomandibular anatomy and patterns of resorption during atrophy. Chapitre 2. Endosseous implants : scientific and clinical aspects. Watsek G, ed. Quintessence Pub Co 1996;29-62.
• HANSSON A, JOHNSSON K, JACOBSSON M, TURESSON I. Removal torques for titanium implants following irradiation. in : Tissue integration in Oral, Orthopedic and Maxillofacial reconstruction. Laney W, Tolman D, eds. Proceedings of the second international congress on tissue integration in oral, orthopaedic and maxillofacial reconstruction. Chicago : Quintessence 1990;228-233.
• HARRISON J. Osteoporosis. In : Metabolic bone disease : Cellular and tissue mechanisms. Tam C, Heersche J, Murray T, eds. Boca Raton : CRC Press, 1989;240-267.
• HUM S, LARSEN P. The effect of radiation at the titanium/bone interface. In : Tissue integration in oral, orthopedic and maxillofacial reconstruction. Laney W, Tolman D, eds. Proceedings of the second international congress on tissue integration in oral, orthopaedic and maxillofacial reconstruction.Quintessence, 1990;234-239.
• JACOBS J, SKIPOR A, BLACK G, URBAN R, GALANE J. Release and excretion of metal in patients who have a total hip-replacement component made of titanium-base alloy. J Bone Joint Surg (am) 1991;73:1475-1486.
• JACOBSSON M, ALBREKTSSON T. Integration of bone implants in a previously irradiated bed. In : Tissue integration in oral and maxillo-facial reconstruction. van Steenberghe D, ed. Proceedings of an International congress. Brussels, 1985. Amsterdam : Excerpta Medica, 1986;110-116.
• JACOBSSON M, TJELLSTROM A, THOMSEN P, ALBREKTSSON T, TURESSON I. Integration of titanium implants in irradiated bone. Histologic and clinical study. Ann Oto Rhinol Laryngol 1988;97:337-340.
• JANDINSKI J, MURRAY P, FENESY K, SCORZIELLO T. Detection of tumor necrosis factor-alpha in crevicular fluid from HIV-infected adults. J Dent Res 1991;70 (special issue): 155 (abstract 1207).
• JAFFIN R, BERGMAN C. The excessive loss of Brånemark fixtures in type IV bone : a 5 years analysis. J Periodontol 1991;62:2-4.
• JEMT T, BOOK K, LINDEN B, URDE G. Failures and complications in 92 consecutively inserted overdentures supported by Brånemark implants in severely resorbed edentulous maxilla : A study from prosthetic treatment to first annual check-up. Int J Oral Maxillofac Implants 1992;7:162-167.
• JEMT T. Implants treatment in elderly patients. Int J Prosthodont 1993;6:456-461.
• JEMT T, LEKHOLM U. Implant treatment in edentulous maxillae : a 5-year follow up report on patients with different degrees of jaw resorption. Int J Oral Maxillofac Implants 1995;10:303-311.
• JISANDER S, GRENTHE B, ALBERIUS P. Dental implant survival in the irradiated jaw : a preliminary report. Int J Oral Maxillofac Implants 1997;12:643-648.
• JOHNSON R, MARX R, BUCKLEY S. Hyperbaric oxygen in oral and maxillofacial surgery. In : Worthington P, Evans J, eds. Controversies in oral and maxillofacial surgery. Philadelphia : Saunders, 1994:107-126.
• JOHNSSON K, HANSSON A, GRANSTRÖM G, JACOBSSON M, TURESSON I. The effect of hyperbaric oxygenation on bone-titanium implant interface strength with and without preceeding irradiation. Int J Oral Maxillofac Implants 1993;8:415-419.
• KELLER J, YOUNG F, HANSEL B. Systemic effects of porous Ti implants. Dent Mater 1985;1:41-42.
• KLUTH E, JAIN P, STUCHELL R, FRICH J. A study of factors contributing to the development of osteoradionecrosis of the jaws. J Prosth Dent 1988;59:194-201.
• KRAUSER J. Nickel-induced allergic contact dermatitis. In : Tissue integration in oral and maxillo-facial reconstruction. van Steenberghe D, ed. Proceedings of an International congress. Brussels 1986. Amsterdam : Excerpta Medica, 1986:490-494.
• KRIBBS P, SMITH D, CHESNUT C. Oral findings in osteoporosis. Part II : Relation between residual ridge and alveolar bone resorption and generalized skeletal osteopenia. J Prosthet Dent 1983;50:719-724.
• KRIBBS P, SMITH D, CHESNUT C, OTT S, KILICOYNE R. Relationship between mandibular and skeletal bone in an osteoporotic population. J Prosthet Dent 1989;62:703-707.
• LALOR P, GRAZ A, WRIGHT S, RAULTON G, FREEMAN M, REVELE P. Sensitivity to titanium. A cause of implant failure. J Bone Joint Surg 1991;37:25-28.
• LARSEN P, STRONCZEK M, LISTON T, MEYERS C. Implant osseointegration in irradiated rabbit tibia with and without hyperbaric oxygen (abstract). Int J Oral Maxillofac Implants 1992;7:125.
• LARSEN P, STRONZECK M, BECK F, ROHRER M. Osseointegration of implants in radiated bone with and without adjunctive hyperbaric oxygen. Oral Maxillofac Surg 1993;51:280-287.
• LUGOWSKI S, SMITH D, McHUGH A, Van LOON J. Release of metal ions from dental implant materials in vivo : Determination of Al, Co, Cr, Mo, Ni, V and T in organ tissue. J Biomed Mater Res 1991;25:1443-1458.
• LYNCH M, JANDINSKI J, HUSAIN A, TIU B, MURRAY PA. Presence of interleukin 1-beta in crevicular fluid from HIV infected adults. J Dent Res 1992;71 (special issue): 155 (abstract 394).
• MAEGLIN B. Kritische stellungnahme zur problematik der zahnärztlichen implantate. In : Schwerpunkte in der oralen implantologie und rekonstruktion. Strub J, Gysi B, Schärer P, eds. Berlin : Quintessence, 1983.
• MANGAROO J, GLASSER J, ROHT L, KAPADIA A. Prevalence of bone demineralization in the United States. Bone 1985;6:135-136.
• MARX R, JOHNSON R. Studies in the radiobiology of osteoradionecrosis and their clinical significance. Oral Surg Oral Med Oral Pathol 1987;64:379-390.
• MISH C. Contemporary Implant Dentistry. St Louis : Mosby, 1993.
• MORI H, MANABE M, KURACHI Y, NAGUMO M. Osseointegration of dental implants in rabbit bone with low mineral density. J Oral Maxillofac Surg 1997;55:351-361.
• NOBLE J. Orthopaedic aspects of osteoporosis. In : Osteoporosis : a multidisciplinary problem. Dixon A, Russell R, Stamp T, eds. London : Royal Society of Medecine, 1983:185-191.
• PAREL S, TJELLSTROM A. The United States and Swedish experience with osseointegration and facial prostheses. Int J Oral Maxillofac Implants 1991;6:75-79.
• RAGNI MV, CROSSET LS, HERNDORN JH. Postoperative infection following orthopaedic surgery in human immunodeficiency virus infected hemophiliacs with CD4 counts < or = 200/mm³. J Arthroplasty 1995;10:716-721.
• RESNICK D, NIWAYAMA G. Osteoporosis. In : Diagnosis of bone and joint disorders. Resnick D, Niwayama G, eds. Philadelphia : Saunders, 1988;2022-2085.
• ROBERTS WE, SIMMONS KE, GARRETTO LP, DECASTRO RA. Bone physiology and metabolism in dental implantology : Risk factors for osteoporosis and other metabolic bone diseases. Implant Dent 1992;1:11-21.
• RYDER MI, WINCKLER JR, WEINREB RN. Elevated phagocytosis, oxidative burst, and factin formation in PMN's from individuals with intraoral manifestations of HIV infections. J Acquir Immune Defic Syndr 1988;1:346-353.
• SANZ M, ETIENNE D. Identification des patients à risque en implantologie orale (I). J Parodont Implant Orale 1998;17:257-272.
• SCHON R, OHNO K, KUDO M, MICHI K. Peri-implant tissue reaction in bone irradiated the fifth day after implantation in rabbits : histologic and histomorphometric measurements. Int J Oral Maxillofac Implants 1996;11:228-238.
• SCHROEDER A. Oral Implantology Basic-ITI Hollow cylinder system. Schroeder A, Sutter F, Krekeler G, eds. Stuttgart : Georg Thieme verlag, 1991:260.
• SENNERBY L, THOMSEN P, ERICSON L. A morphometric and biomechanic comparison of titanium implants inserted in rabbit cortical and cancellous bone. Int J Oral Maxillofac Implants 1992;7:62-71.
• SIXOU M, LODTER JP, DUFFAUT-LAGARRIGUE D. Risque d'endocardite infectieuse à partir de bactéries commensales d'identification difficile de la flore sous-gingivale. Inf Dent 1993;6:337-341.
• SMATT V, ROBIN M, GIBEILI Z, HADJALI C, VANZO L. Approche thérapeutique implantaire dentaire en terrain irradié. Rev Stomatol Chir maxillofac 1995;96:396-400.
• SMITH R, BERGER R, DODSON T. Risk factors associated with dental implants in healthy and medically compromised patients. Int J Oral Maxillofac Implants 1992;7:367-372.
• SMITH D, LUGOWSKI S, McHUGH A, DEPORTER D, WATSON P, CHIPMAN M. Systemic metal ion levels in dental implant patients. Int J Oral Maxillofac Implants 1997;12:828-834.
• TAYLOR T, WORTHINGTON P. Osseointegrated implant rehabilitation of the previously irradiated mandible : results of a limited trial at 3 to 7 years. J Prosthet Dent 1993;69:60-69.
• TOLMAN D, TAYLOR P. Bone-anchored craniofacial prosthesis study : irradiated patients. Int J Oral Maxillofac Implants 1996;11:612-619.
• UEDA M, KANEDA T, TAKAHASHI H. Effect of hyperbaric oxygen therapy on osseointegration of titanium implants in irradiated bone : A preliminary report. Int J Oral Maxillofac Implants 1993;8:41-44.
• UNGER AS, KESSLER CM, LEWIS RJ : Total knee arthroplasty in human immunodeficiency virus-infected hemophiliacs. J Arthroplasty 1995;10:448-452.
• VAN STEENBERGHE D. Periodontal aspects of osseointegrated oral implants Modum Branemark. Dent Clin N Amer 1988;32:355-370.
• WAHL M. Myths of dental-induced endocarditis. Compend Contin Educ Dent 1994;15:1100-1119.
• WANG R, PILLAI K, JONES P. In vitro backscattering from implant materials during radiotherapy. J Prosthet Dent 1996;75:626-632.
• WATZINGER F, EWERS R, HENNINGER A, SUDASCH G, BABKA A, WOELFL G. Endosteal implants in the irradiated lower jaw. J Craniomaxillofac Surg 1996;24:237-244.
• WESTFELT E, RYLANDER H, BLOHMÉ G, JONASSON P, LINDHE J. The effect of periodontal therapy in diabetics. results after 5 years. J Clin Periodontol 1996;23:92-100.
• WINCKLER JR, HAMMERLE C. PMN function in HIV infected men-effect of endotoxin and serum. J Dent Res 1991;70 (special issue) : 537 (abstract 2171).
• WINCKLER JR, ROBERTSON PB. Periodontal disease associated with HIV infection. Oral Surg Oral Med Oral Pathol Med 1992;73:145-150.
• ZACKIN J. Bacterial endocarditis following periodontal maintenance with prophylaxis. Periodontal Insights 1997;4:7-9.