Routine Prophylactic Antibiotic Use in Diabetic Dental Patients

Copyright 1999 Journal of the California Dental Association.

There is no scientific evidence in the literature to support the premise that well-controlled, or even moderately well-controlled, nonketotic diabetic patients are prone to infection when undergoing uncomplicated dentoalveolar surgery. Routine administration of prophylactic antibiotics should be considered only in situations where prophylactic antimicrobials would be used for a nondiabetic patient. Poorly controlled diabetics (whether Type I or II), with fasting glucose levels above 250 mg/dL, should be referred for improved control of their blood sugar before nonemergency surgery is performed. If emergency surgery is needed for a poorly controlled patient, then prophylactic antibiotics are prudent, using the accepted principles of such use. Infections in diabetic patients, regardless of their control levels, should be managed aggressively, including possible early referral to oral and maxillofacial surgeons.

Unproven myths abound in the fields of dentoalveolar surgery and surgical pharmacology. A number of these myths were recently explored and shown to be scientifically unproved or illogical.¹ Another, similar, area of confused clinical guidance relates to the prophylactic use of antibiotics following surgery in diabetic patients. For many practitioners, the guidance provided in the past has been a very vague and generalized axiom that seemingly has evolved over the years. It can be paraphrased like this: "All diabetics, as a group, are more prone to infection and therefore should have prophylactic antibiotics routinely prescribed for all dental surgery." Zoeller and Kadis present this viewpoint as their interpretation of the consensus of the current literature.² It is the purpose of this paper to examine the validity of this basic premise.

Diabetes and the Immune System

It is estimated that there are 15 to 20 million Americans (2 percent to 4 percent of the population) who have one form of diabetes mellitus or another.³ The prevalence has increased significantly during the past 40 years. One of the least common forms is the so-called Type I diabetes, also referred to as insulin-dependent diabetes or "juvenile-onset" diabetes, which affects only 5 percent of the diabetic population. The vast majority of cases are the so-called Type II form, also referred to as "adult-onset" diabetes or non-insulin-dependent diabetes mellitus (even if the patient is using insulin). There are other variants of the disease; and the reader is referred to current textbooks, such as Little and colleagues,³ for further information on this family of diseases.

Does diabetes mellitus, indeed, consistently disrupt immune system performance? There are numerous studies and anecdotal reports in the literature on the subject of susceptibility of diabetic patients to infection, but the common denominators in many of those articles suggest the problem is confined to a fairly narrow set of circumstances:

* The patients are largely Type I diabetics who are in poor control and physiologically unstable.

* The infections often involved one or more extremities (usually lower) with notably poor vascular supply.

* The infections studied were pre-existing, requiring therapeutic management (not prophylaxis); few articles address the aspect of giving antibiotics to prevent infections (other than periodontal disease).

* Very few of these reported cases involved oral infections, other than periodontal disease, or involved head and neck infections in well-controlled diabetic patients.

* Many articles are filled with assumptions and contradictions, making them difficult to interpret.⁴

It is also unclear in many cases whether poor metabolic control was a contributing cause of the infection or actually resulted from the infection.

Overview

The defining of altered host responses in diabetes is hampered by the complexity of the immune system and diabetes itself. In vivo, the various "arms" of the immune system are highly dynamic and interdependent.⁵ It is overly simplistic to evaluate any single element of the system in isolation and equally difficult to compare studies.⁵ The literature is replete with contradictory findings, inconclusive results, and disagreement about the ability of the immune system to function normally in well-controlled diabetics. This is a reflection of the heterogenous nature of diabetes and the fact that the inclusion criteria for participants in many studies are often not well-defined. The literature more consistently suggests multiple immune system compromises in a small subset of unstable diabetic patients who exist in poor metabolic states with poor glucose control. Even in those patients, however, the clinical significance of impaired neutrophil function has not been fully determined.⁴ Other studies have shown that granulocyte adherence, chemotaxis, phagocytosis, and microbicidal function in patients whose blood sugar levels are aggressively controlled are improved.⁶ Unquestionably, glycemic control is important for the enhancement of leukocyte function.

A Medline search of the literature over the past 25 years has failed to uncover any valid studies in which the susceptibility of relatively well-vascularized, oral wounds (such as extraction sites) to acute infections was examined in well-controlled diabetic patients as opposed to nondiabetic patients, except for studies relating to periodontal disease.

Blood Supply

Maintenance of normal oxygenation and nutrition to the tissues and continuous delivery of humoral and cellular components of the immune system to the site(s) are dependent on an adequate blood supply to those tissues.⁴ Diabetes is known for its cumulative damage to the microvasculature. Patients with infections in areas of poor vascular supply will not be able to respond to them with the same intensity as a noncompromised patient. In a poorly vascularized extremity, this is a clinically significant problem. With the relatively ample vascularity of the head, neck, and oral cavity, however, vascular compromise has not been shown to be a relevant factor.

Poor blood supply has also been shown to alter cellular components, increase local acidosis, and increase vascular permeability.⁴ Interestingly, compromised blood supply to the tissues can also inhibit delivery of an antibiotic to tissue sites, an issue that is rarely addressed by those advocating routine use of prophylactic antibiotics.

Humoral Immunity

Although some earlier studies suggested otherwise, more recent studies show that diabetic patients respond as well as control patients to vaccinations.⁴ No correlations between antibody response and patient age, glucose levels, or duration of disease have generally been shown, although one study shows that elderly patients do not respond as well to pneumococcal vaccines and that their antibody levels decline more rapidly.⁴ The majority of studies of serum complement in diabetics have found normal or elevated levels.⁴ Levels of antibodies against specific microorganisms (such as Pneumococcus) are no different in diabetics than in nondiabetics.⁷ Complement deficiency is common in diabetics, but the clinical implications are unclear.⁷

Phagocytic Function

The inconsistent data in the literature are testimony to the difficulty in evaluating phagocytosis, since there are so many different steps in the process. Antigens must be sensitized by antibodies (opsonization), and phagocytes must be able to migrate to the area of infection (chemotaxis) and penetrate the endothelium of capillary walls (diapedesis), and then operate in the acidic environment of the infected tissue(s).

In vitro studies have suggested delayed chemotaxis in both Type I and Type II diabetic patients. Other studies suggest that diabetics may have abnormal chemotactic responses as part of the genetic makeup of their polymorphonuclear leukocytes.⁵ Decreased phagocytosis is especially notable when fasting blood glucose levels are greater than 250 mg/dL.⁵ On the other hand, one computer-enhanced chemotaxis study has shown that a diabetic’s chemotactic cells move at normal rates.⁸

Defective engulfment and intracellular killing by phagocytes in diabetic patients has been reported in several studies, but in studies of diabetic patients in which Staphylococcus aureus was used, ingestion of microorganisms was only found to be abnormal in patients who were uncontrolled and in ketoacidosis. Other studies show that a phagocytic cell’s ability to mount an oxidative attack is reduced in the presence of high glucose levels.⁷ Sentochnik and Eliopoulos cite a study that demonstrates a defect in phagocytosis of S. aureus in patients with Type II diabetes but provided no correlation with the adequacy of glycemic control.⁵

Lymphocytic Action

Again, the literature is contradictory. The lymphocyte response to Candida antigen is reported to be normal in diabetic subjects in one study, but another study’s results do not support that finding in patients who are in poor control.⁴ This impairment normalizes with the reinstitution of good metabolic control. In a study in the 1970s, Gilbert and associates found that diabetics exhibited basal levels of lymphocytes comparable to nondiabetics and that patients initial response to administration of an endotoxin was identical in both groups, but diabetic patients had reduced levels of circulating humoral elements after the third day.⁹ The authors stress that the physiologic significance of those findings was undetermined, however.

Periodontal Disease

It has been shown in numerous studies that periodontal disease seems to be more common and more severe in diabetic patients than in nondiabetics.^4,10-12 In one animal study, 48 percent of the diabetic animals were shown to have impaired leukotaxis in the gingival crevices, and several had increased numbers of anaerobic microorganisms.⁴ Nevertheless, numerous international studies on the relationship of diabetes and periodontitis have reported varying results.¹² Studies in the United States and other countries consistently fail to find overall differences in the prevalence of periodontal pockets, alveolar bone loss, or tooth loss in diabetic patients when compared to age-matched nondiabetics.¹² These conflicts are noted to be likely due to variations in the types of diabetes, severity, control, duration, and differences in oral conditions among patients. There has been at least one case report of a severe deep neck infection originating from a periodontal abscess. This appears to be an uncommon sequela, however; and it, too, occurred in an uncontrolled, ketotic diabetic.

Patients with controlled diabetes generally respond as well to periodontal treatment as nondiabetics.¹² It has also been shown that Type I patients whose disease is under strict metabolic and clinical control have periodontal complications at a frequency comparable to nondiabetic patients.¹³ It further remains unclear and unproved whether a patient’s susceptibility to chronic periodontal disease has any validity or relationship to a patient’s susceptibility to acute infection following the performance of other dentoalveolar surgery, including extractions, in well-controlled and moderately well-controlled, Type I and Type II diabetic patients.

Response to Infections

Several studies have demonstrated that diabetics cope poorly with staphylococcal and Candida infections of the skin, and one study demonstrates that abscesses persist for longer periods in diabetic mice.⁴ Indeed, staphylococcal infections of the skin are twice as common in diabetic patients as in nondiabetic patients with other disabling diseases.⁴ Many extremity studies are also complicated by the frequent presence of osteomyelitis of the contiguous bones. The consensus of the literature is that studies of extremity infections in diabetic patients support the concept that compromised peripheral oxygen supply in the presence of an impaired peripheral vascular system and neuropathy contributes more to delayed healing and onset and establishment of infections than any other factors.⁴ Again, there are few data that this is also a concern in the relatively well-vascularized oral environment.

Diabetics are thought to have an increased incidence of oral candidiasis,¹⁴ but Fisher and associates found no correlation between level of sugar control and yeast colonization.¹⁵ A study of diabetics and nondiabetics with dentures fails to reveal any statistically significant difference in the incidence of denture stomatitis.¹⁶ Other studies, however, did find more mucosal colonization in diabetic patients, but some studies were flawed in their design and many originated in overseas countries where conditions may not equate to those in the United States.⁵

Postsurgical Wound Infections

One article has estimated that approximately 50 percent of the general population that has diabetes mellitus will require at least one operation during their lifetimes, and approximately two-thirds of those patients will experience infectious complications.¹⁷ There is no reliable data, however, on how many diabetics undergo dental surgery and experience clinically significant odontogenic infections during their lifetime.

A large-scale study was carried out on 23,649 postoperative general surgery patients (both diabetic and nondiabetic), and it was found that the clean-wound infection rate in diabetics was more than five times greater than in the general population (10.7 percent as opposed to 1.8 percent).¹⁸ Babineau and Bothe note that the study was criticized because it did not take other risk factors into consideration, such as advanced age, nutritional status, levels of control, and co-morbid diseases.¹⁷ In another study of 100 patients undergoing elective general surgery, diabetic patients with elevated glucose levels on postoperative day 1 had a 2.7 times greater infection rate than diabetic patients with more normal glucose levels.⁶ In general, however, it is well-accepted that the incidences of infections and wound healing are similar between well-controlled diabetic and nondiabetic patients, except for extremity procedures.⁷ In a retrospective study of 9,000 general surgery patients, the primary risk factor for postoperative infection was the presence of cardiac failure or valvular heart disease, which hampered patient mobility.¹⁹

The literature, therefore, appears to support the concept that when infections occur, they may be more severe and protracted in poorly controlled diabetics, perhaps due to impaired leukocyte function in later days of the infection process and/or compromised blood supply.⁹

It is known that severe surgical stress is accompanied by a marked increase in plasma glucagon, epinephrine, and cortisol. Counterregulatory hormones then increase hepatic glucose release and decrease glucose cellular intake, resulting in a relative hyperglycemic state. This effect appears to be more exaggerated in the diabetic patient than in the nondiabetic.⁶ It is further known that certain bacteria thrive in a hyperglycemic state. Studies cited by McMurry suggest that gram-positive bacteria (including staphylococcus) thrive in hyperglycemic serum while gram-negative bacteria grow less well, which may partially explain some of the observations that diabetic patients are prone to infections.²⁰ So, the question is this: How much stress is caused by uncomplicated outpatient dentoalveolar surgery, and is the severity of that surgically induced stress sufficient to trigger that physiologic state? This question has not been adequately addressed in the literature.

Wound Healing

Despite difficulties in interpreting data in the literature, the consensus is that poorly controlled diabetics do have poor wound healing, but adequate control with insulin usually resolves the problem.²⁰ Data also suggest that increased age and obesity make wounds increasingly prone to infection; but age and obesity are also factors in the development and advancement of the diabetic disease process itself, so it is difficult to separate the effects of one from the other. McMurry cites a study that indicates the percentage of wound infections in diabetic patients is more than double that of nondiabetic patients, but when the data are adjusted for age, the incidences of wound infection are nearly identical.²⁰ Studies have shown delayed wound healing, poor collagen formation, and poor tensile wound strength in diabetic animals; but these are corrected with restoration of adequate control (i.e., insulin administration).²⁰

Optimal Glucose Control

It appears that an optimal blood glucose level for patients undergoing surgery is between a level where the surgical patient is not at risk for hypoglycemic emergencies and a level where wound healing and granulocytic function are not impaired. This level has not been precisely defined, but various authors suggest that the patient should have a serum glucose level at or slightly below 180 to 250 mg/dL.^7,10,20,21

Discussion

It is widely believed that patients with diabetes mellitus are more prone to infection and other postsurgical complications, and all such patients require routine antibiotic prophylaxis for dental procedures. The professional literature fails to support this premise.

This myth of increased susceptibility to infection in diabetic patients following dentoalveolar surgery may have evolved from misapplication of information in the medical literature that documents severe extremity infections in diabetic patients. Impaired vascularity often predisposes the peripheral tissues in extremities to infection. In the presence of ongoing tissue hypoxia and impaired perfusion in poorly controlled diabetics, any pre-existing immune deficiencies are exaggerated by the presence of compromising neuropathy, which can lead to lack of patient attention to extremity skin wounds, malnutrition, co-morbid disease processes, and difficulty getting systemic antibiotics to the peripheral infection site. There is no reliable evidence, however, that relatively well-vascularized oral wounds are equally susceptible, nor is there any evidence that well-controlled Type I and Type II diabetic patients are at increased risk for postsurgical oral infections (other than chronic periodontal disease).

Many older studies, upon which this myth may rest, derived data from autopsy investigations following infections of the urinary tract, respiratory tree, and extremities; and controls were typically lacking.⁵ More recent studies have attributed the complications and increased mortality to cardiovascular disease rather than uncontrolled infection.⁵

Babineau and Bothe note that recent well-designed clinical studies have shown that well-controlled diabetes (Type I or II) is no longer a risk by itself for postoperative surgical complications.¹⁶ In retrospective studies, diabetes was not proved to be an independent risk factor for complications from vascular, abdominal, or hip surgery, whether the patient was insulin -- or orally -- controlled.^19,22

Nevertheless, surgical intervention still creates an increased sense of anxiety in both the doctor and the patient. Diabetic patients, especially Type I, continue to represent a subset of patients that are characterized as high risk. Authors emphasize that it is essential that the patient’s blood sugar levels be optimized prior to surgery. Although these diabetic oral surgery patients are not necessarily more susceptible to wound infections, infections in that population can be more severe and prolonged than in nondiabetic patients.^8,17,23 The question is, therefore, this: If we accept the position that well-controlled (Type I or II) diabetics are not more susceptible to postoperative infections, does the impairment of immune response when infection do occur justify the routine prophylactic administration of an antibiotic in well-controlled or moderately well-controlled diabetic patients following all types of dentoalveolar surgery? Generally, the answer is no.

Dentoalveolar Surgery

Pedersen and, more recently, Alling and colleagues have forwarded the position that well-controlled diabetic patients do not require prophylactic antibiotic therapy for routine oral surgical procedures, and delayed wound healing should not be anticipated in the rich vascular environment of the oral cavity.^24,25 Diabetics who are poorly controlled, however, may be at increased risk and therefore suitable candidates for administration of prophylactic antibiotics. If used, prophylactic antibiotics should be administered prior to the surgical procedure(s) and for a short duration, in accordance with currently recommended usage principles. It is beyond the scope of this paper to review the appropriate application of antibiotic principles, and the reader is referred to other articles for guidance.^26,27 Generally, antibiotics would be used in a protocol similar to that of the American Heart Association for prevention of infective endocarditis.²⁸ Starting antibiotic regimens after the surgery is completed is not considered an appropriate methodology for prophylaxis.^26,27

More importantly, diabetic patients should be brought under proper control prior to elective surgery. That step by itself will reduce their risk for infection. It is not necessary to make the patient normoglycemic; as noted, glucose levels at or below 250 mg/dL appear to be acceptable. Levels below 100 mg/dL may incur a risk for hypoglycemic emergencies and therefore should be avoided.²⁰ In the absence of adequate glycemic control, initiation of prophylactic antibiotic coverage is prudent before the elective surgical intervention, using recognized protocols.

With this in mind, it is very useful for every dental office to have a state-of-the-art fingerstick glucometer to allow accurate chairside glucose testing. These now-automated devices are simple to use, reasonably priced, readily available in any pharmacy, and provide highly accurate readings in 30 to 60 seconds. This is a valuable tool that helps clarify a patient’s status and facilitates prudent clinical decisions.

The use of 0.2 percent chlorhexidine gluconate may have some benefit as a presurgical rinse. One study demonstrated that exposure to chlorhexidine gluconate for one minute reduced colonies of Candida in the buccal mucosae,²⁹ but it is unknown whether preoperative chlorhexidine gluconate use in diabetic patients has any significant impact on reducing the incidence of post-surgical infections. Given the lack of any significant adverse effects, chlorhexidine gluconate use might be worthy of consideration, despite the lack of evidence of clinical efficacy.

Acute Infection Management

When orofacial infection is diagnosed in a diabetic patient, whether well-controlled or not, it should then be treated aggressively. Management of acute odontogenic infections is beyond the scope of this manuscript, but guidance can be found in numerous textbooks and articles in the contemporary literature.^24,27 Management typically consists of:

* Administration of appropriate antibiotics;

* Early surgical drainage of pus;

* Adequate hydration and nutrition; and

* Referral to an oral and maxillofacial surgeon.

In the diabetic patient, the additional aspect of frequent glucose testing and aggressive glycemic control would be essential. Close coordination with the patient’s internist or endocrinologist is imperative when dealing with severely infected diabetic patients, so adequate control of the patient’s metabolic state can be maintained.³⁰ Frequently, such patients require admission to a hospital and inpatient management. Antibiotic selection would be guided by the same factors affecting the choices for any other patient. Further elaboration on antibiotic selection is beyond the scope of this paper, and the reader is referred to other references for more information.^24,26,27,31

Conclusions

There is no scientific evidence in the literature that well-controlled, nonketotic, diabetic patient are more prone to infection than nondiabetic patients when undergoing surgery. Once infected, however, they may have a more severe and prolonged clinical course. Routine administration of prophylactic antibiotics is not necessary in situations where antibiotics would not be considered for a nondiabetic patient, however. Poorly controlled diabetics, with fasting glucose levels consistently above 250 mg/dL, should be referred for improved control of their blood sugar before nonemergency surgery is considered. If surgery is essential in a poorly controlled patient, then prophylactic antibiotics are indicated, following accepted principles of such use. Infections in diabetic patients, regardless of their control levels, should be managed aggressively, including possible early referral to oral and maxillofacial surgeons.

Acknowledgment/The author expresses his grateful appreciation to Dr. Robert G. Triplett, chairman, Department of Oral and Maxillofacial Surgery and Pharmacology, Baylor College of Dentistry, the Texas A&M University System Health Science Center, for his review of this manuscript and constructive suggestions.

Author/

Roger E. Alexander, DDS, is an associate professor in the Department of Oral & Maxillofacial Surgery & Pharmacology at Baylor College of Dentistry, the Texas A&M University System Health Science Center.

References/

1. Alexander RE, Eleven myths of dentoalveolar surgery. J Am Dent Assoc 129:1271-9, 1998.

2.Zoeller GN, Kadis B, The diabetic dental patient. Gen Dent 29:58-61, 1981.

3. Little JW, Falace DA, et al, Dental Management of the Medically Compromised Patient, 5th ed. Mosby, St. Louis, 1997, p 387.

4. Currie BP, Casey JI, Host defense and infections in diabetes mellitus. In, Porte D, Sherwin RS, eds., Ellenberg and Rifkin’s Diabetes Mellitus, 5th ed. Appleton & Lange Publishers, Stamford, CT, 1997, pp 861-71.

5. Sentochnik DE, Eliopoulos GM, Infection and diabetes. In, Kohn CR, Weir GC, eds, Joslin’s Diabetes Mellitus, 13th ed. Lea & Febiger, Philadelphia, 1994, pp 867-70, 878, 882.

6. McMahon MM, Bistrian BR, Host defenses and susceptibility to infection in patients with diabetes mellitus. Infect Dis Clin N Am 9:1-9, 1995.

7. Pickup J, Williams G, eds, Textbook of Diabetes, Vol 2. Blackwell Scientific, Oxford, 1991, pp 813-6 and 820-5.

8. Donovan RN, Goldstein E, et al, A computer-assisted image-analysis system for analyzing polymorphonuclear leukocyte chemotaxis on patients with diabetes mellitus. J Infect Dis 155:737-41, 1987.

9. Gilbert HS, Rayfield EJ, et al, Effects of acute endotoxemia and glucose administration on circulating leukocyte populations in normal and diabetic subjects. Metabolism 27:889-99, 1978.

10. Shlossman M, Knowler WC, et al, Type 2 diabetes mellitus and periodontal disease. J Am Dent Assoc 121:532-6, 1990.

11. Scully C and Cawson RA, Medical Problems in Dentistry. Wright, Oxford, England, 1993, pp 279-82.

12. Oliver RC, Löe H, Diabetes and oral diseases. In, Porte D, Sherwin RS, eds, Ellenberg and Rifkin’s Diabetes Mellitus, 5th ed. Appleton & Lange, Stamford, CT, 1997, pp 1227-33.

13. Pinducciu G, Micheletti L, et al, Periodontal disease, oral microbial flora and salivary antibacterial factors in diabetes mellitus type I patients. European J Epidem 12:631-6, 1996.

14.Ueta E, Osaki T, et al, Prevalence of diabetes mellitus in odontogenic infections and oral candidiasis: an analysis of neutrophil suppression. J Oral Path Med 22:168-74, 1993.

15. Fisher BM, Lamey PJ, et al, Carriage of Candida species in the oral cavity in diabetic patients: relativity to glycaemic control. J Oral Path 16:282-4, 1987.

16. Phelan JA, Levin SM, A prevalance study of denture stomatitis in subjects with diabetes mellitus or elevated plasma glucose levels. Oral Surg Oral Med Oral Pathol 62:303-5, 1986.

17. Babineau TJ, Bothe A, General surgery considerations in the diabetic patient. Infect Dis Clin N Am 9:183-93, 1995.

18. Cruse PE, Foord R, A five-year prospective study of 23,649 surgical wounds. Arch Surg 107:206-10, 1973.

19. MacKenzie CR, Charlson ME, Assessment of perioperative risk in the patient with diabetes mellitus. Surg Gynec Obstet 167:293-9, 1988.

20. McMurry JF, Wound healing with diabetes mellitus. Surg Clin N Am 64:769-77, 1984.

21. Pendergrass M, Graybill J, Infections and Diabetes Mellitus. In, DeFronzo RA, ed, Current Therapy of Diabetes Mellitus. Mosby, St. Louis, 1998, pp 218-23.

22. Hjortrup A, Sorensen C, et al, Influence of diabetes mellitus on operative risk. Br J Surg 72:783-5, 1985.

23. Morain WD, Colen LlB, Wound healing in diabetes mellitus. In, Miller SH, Rudolph R, eds., Cl Plast Surg 17:493-8; 1990

24. Pedersen GW. Oral Surgery. WB Saunders, Philadelphia, 1988, pp 110-1.

25. Alling CC, Helfrick JF, Alling RD, Impacted Teeth. WB Saunders, Philadelphia, 1993, p 86.

26. Pallasch TJ, Pharmacokinetic principles of antimicrobial therapy. Periodontol 2000 10:5-11, 1996.

27. Alexander RE, Basic principles of antibiotic therapy and prophylaxis. Quintessence Internat 28:815-25, 1997.

28. Dajani AS, Taubert KA, et al, Prevention of bacterial endocarditis: recommendations by the American Heart Association. J Am Med Assoc 277:1794-801, 1997.

29. Darwazeh AM, Lamey PJ, et al, The effect of exposure to chlorhexidine gluconate in vitro and in vivo on in vitro adhesion of Candida albicans to buccal epithelial cells from diabetic and non-diabetic subjects. J Oral Path Med 23:130-2, 1994.

30. Hall EH, Sherman RG, et al, Antibacterial prophylaxis. Dent Clin N Am 38:707-18, 1994.

31. Thompson RL, Wright AJ, General principles of antimicrobial therapy. Mayo Clin Proc 73:995-1006, 1998.

To request a printed copy of this article, please contact/Roger E. Alexander, DDS, Baylor College of Dentistry, P.O. Box 660677, Dallas, TX 75266-0677.