A 29-year-old asplenic male with overwhelming sepsis after tooth extractions and postoperative infection

EVIDENCEBASED CASE CONFERENCE

CASE PRESENTATION

A 29-year-old asplenic male with overwhelming sepsis after tooth extractions and postoperative infection Mark Drangsholt, DDS, MPH,a,b Mark M. Schubert, DDS, MSD,a,c Christine S. Ritchie MD, MSPH,d From the Departments of Oral Medicinea and Dental Public Health Sciences,b School of Dentistry, University of Washington, Seattle; the Seattle Cancer Care Alliancec; and the Department of General Internal Medicine,d University of Louisville, Louisville, Ky A 29-year-old male carpenter was first admitted to a hospital complaining of a 24-hour episode of nausea, vomiting, diarrhea, fever, and chills. He had been well and working until 18 days earlier, when he had seen a general dentist because of upper right tooth pain.

HISTORY The dentist performed an examination; took periapical radiographs of tooth #2 and root tips #4 and #19 (Fig 1); diagnosed the patient with nonrestorable teeth #2, #4, #19, presumably due to dental caries; and referred him to an oral and maxillofacial surgeon. The patient was seen 3 days later by the surgeon, who subsequently surgically extracted tooth #2 and root tips #4 and #19 on the same day, after putting the patient under conscious sedation and local anesthesia with midazolam, meperidine, Phenergan, and mepivicaine. This surgery was performed 15 days prior to the patient’s admission to the hospital. The medical history that was recorded at that time did not show that the patient had previously undergone a trauma-related splenectomy at age 16, but the anesthesia record completed partially by an assistant revealed, under the category of “prior surgeries,” that

J Evid Base Dent Pract 2003;3:156-64 © Mosby, Inc. All rights reserved. doi:10.1067/med.2003.62

156

the patient had had his spleen removed. After the tooth extractions were completed, he was given a 5-day course of penicillin, 500 mgqid. Over the next few days, the patient noticed that the pain did not subside in the upper right area of his mouth; instead he noted more tenderness and swelling in the area where tooth #2 had been previously located. After 7 days, he contacted the oral surgeon’s office and was seen for a postoperative check, noting that he had continued pain and a bad taste in his mouth. No abnormalities were noted during a brief exam by an assistant, and he was not seen by the surgeon. No additional therapeutic measures were recommended. His extraction–site-related pain did not resolve. Eight days later he presented to the emergency room with fever, mottled skin, dyspnea, and tachycardia. Initial vital signs revealed a respiratory rate of 35, pulse of 124, blood pressure of 110/61, and temperature of 104ºF. His posterior pharynx appeared to be swollen and inflamed and show signs of tonsillitis, which was later reinterpreted as a sign of his septic systemic condition. Bowel sounds were reduced, and his abdomen was tender. His extremities were mottled and cool. A chest radiograph showed no acute infiltrates; white

Fig 1. Periapical radiographs of right maxillary molars and left mandibular molars. These copies of the original radiographs show deep occlusal caries on tooth #2 and only root tips on tooth #4 and #19. No obvious periapical radiolucency is apparent on the apices of tooth #2, although the x-ray quality is poor. Both #4 and #19 root tips appear to have two small periapical radiolucencies.

blood cell count was 5200; hematocrit was 47%; and room air blood gas was notable for a marked respiratory alkalosis with a pH of 7.42, PO2 of 80, and PCO2 of 17.5. He was admitted with a presumptive diagnosis of postsplenectomy sepsis probably due to encapsulated bacteria such as Streptococcus pneumoniae, Neisseria menigiditis, or Hemophilus influenzae. The patient’s status quickly deteriorated and he was resuscitated with intravenous fluids. Swan-Ganz and arterial lines were placed. After cultures were taken, he was started on intravenous cefotaximine and clindamycin. He was also given gamma globulin and continued on supplemental oxygen. Subsequent laboratory results showed prolonged coagulation times consistent with disseminated intravascular coagulation. He was given platelets, cryoprecipitate, packed red blood cells, and fresh frozen plasma. Two days later, cultures grew S pneumoniae sensitive to penicillin, ampicillin, ceftriaxone, and ciprofloxacin, so his antibiotics were changed to ampicillin and ceftriaxone. His renal function deteriorated, presumably secondary to acute tubular necrosis, and he was transferred to another hospital for possible renal dialysis. He subsequently developed gangrene in all of his extremities, and required both left and right amputations below the knee as well as amputation of all of his digits 6 weeks later. In the following 2 months, he underwent plastic reconstructive surgery of his hands and lower back. He was discharged on day 131, after 4 and a half months of hospitalization and was later fitted with prosthetic lower limbs on an outpatient basis. Journal of Evidence-Based Dental Practice Volume 3, Number 3

DISCUSSION Infections in people without spleens Overwhelming postsplenectomy infection (OPSI) is a serious and often rapidly fatal condition that can develop in persons without a functional spleen.1 The spleen is a fist-sized, highly vascularized spongy organ situated in the upper left abdomen. Besides controlling the quality of circulating erythrocytes, it is a vital component in the phagocytosis of microorganisms and subsequent antibody production.2 Encapsulated bacteria— especially S pneumoniae, N meningitidis, and H influenzae, which are usually cleared from the blood by the spleen—have been shown to cause about 80% of these severe infections. However, there are indications that this bacterial spectrum may be changing with the advent of bacterial vaccines and may now include Esherichia coli, Pseudomonas aeruginosa, and Captocytophaga.3,4 The case fatality rate has dropped, but is still approximately 40% to 50%,5 making OPSI an extremely dangerous condition to develop. The risk for this severe complication, which may be higher in children, has been shown to be greatly elevated up to 30 days after splenectomy, with continued elevation for the first 2 years after spleen removal.6 Patients experience a lifelong increased risk of severe sepsis, which occurs in 0.23% to 0.42% of all splenectomized patients per year,5,7,8 and is many times higher than in a person with a functioning spleen. The lifetime risk of overwhelming infection in asplenic persons is estimated to be about 5%.9 Most cases of OPSI do not show a primary focus of Drangsholt, Schubert, and Ritchie 157

TABLE I. Determining whether a dental procedure or condition caused harm: causal criteria applied to case reports Type of study

Randomized controlled trial (best) Cohort Case-control Cross-sectional Case-series Case-report (worst) Magnitude of association Common condition or procedure + rare harmful outcome + (if only case reports exist) close association in time† Temporal sequence Procedure or condition definitely occurred before the harmful problem Consistency Other case reports of similarly associated dental procedures and problems have been reported Biological plausibility For infectious problems, organism that was associated with the problem is a known oral organism Analogy Similar medical procedures have been associated with similar problems Dose response More intensive procedure or extensive condition is higher risk Reversibility* Removal of exposure results in reversal of condition back to original state Specificity Only specific types of dental procedure or disease causes the problem *Newer aspect of causation that only applies for some exposures such as drugs. † Association in time corresponds to the known induction period (time from dental procedure to first signs and symptoms of the harmful condition.

infection,1 and the risk factors for the condition are not well characterized. It has been suggested that bacteremia-producing dental procedures could possibly precipitate OPSI,10 although there is no controlled evidence to support this hypothesis. The prevalence of asplenic individuals in the United Kingdom is about 1 per 1000, or 0.01%,11 and is probably similar in the United States. To our knowledge, this is the first case of an OPSI associated with prior dental extractions and a subsequent postoperative infection. Undoubtedly, the key question is whether the dental treatment or related postoperative infection caused this condition.

Did a dental procedure and or postoperative infection cause the postsplenectomy sepsis? Answerable clinical question. When a poor outcome like death or severe infection occurs shortly after a dental procedure, the question of whether that procedure caused the problem arises. This answerable clinical question, written in PICO (Patient, Intervention, Comparison, Outcome) format,12 could be, “In persons without functional spleens (patient), do dental procedures or infections (interventions/exposures) cause overwhelming and life-threatening sepsis (outcome) compared (comparison) with people without such procedures?” This question is one of etiology, or harm, and to answer such questions, we need to evaluate evidence about causation.13 The ideal study design to address this ques158 Drangsholt, Schubert, and Ritchie

tion would be a randomized controlled trial of asplenic persons with or without dental extractions performed to treat their deep carious lesions and a subsequent assessment of the number of postsplenectomy infections in each group. Since it is not only infeasible but also unethical to allocate thousands of people either to deep carious lesions or extractions, the next best study design would be a large cohort study of splenectomized persons. They would be followed up until they developed a serious infection or until the study time ran out. Then the baseline factors (dental procedure or not) could be compared with whether a person developed a serious infection, while controlling for other important factors. Although such studies have recently been done to evaluate the value of vaccinating people against S pneumoniae,3 they have not been conducted to evaluate the risk of either dental procedures or dental infections. No case-control nor cross-sectional studies have been published either. Only this case report and several others with oral bacteria have been published (see article searches that follow). Usually, case reports can only raise hypotheses and alert clinicians and researchers to possible relationships between exposures and diseases, but, in this case, for both legal and clinical reasons, an answer was needed. In this section, we would like to address the accepted criteria for causation one by one to evaluate the most likely explanation for the present case of overwhelming postsplenectomy infection. The Journal of Evidence-Based Dental Practice September 2003

TABLE II. Criteria for causation applied to competing hypotheses of the cause of the sepsis Magnitude of association Rank order

Competing hypotheses for cause of sepsis in the present case

Exposure

1

Poorly healing and infected extraction site #2 in close unknown proximity to the maxillary sinus created a S pneumoniae bacteremia 2 Small oroantral fistula created after surgical extraction #2 uncommon allowed the resultant infection and bacteremia with S pneumoniae 3 Occult maxillary sinusitis with S pneumoniae uncommon 4 Spontaneous S pneumoniae bacteremia without oral or rare? sinus focus 5 Tonsillitis with S pneumoniae uncommon 6 S pneumoniae bacteremia from periapical abscess #4, rare #19 (#2?) 7 S pneumoniae bacteremia from dental extraction of #2, non-existent #4, #19 8 Other uncultivated organisms caused the OPSI unknown NA Dental extraction and S sanguis bacterial endocarditis uncommon

Outcome

Close in time

Induction time

rare

yes

< 8 days

rare

yes

8 days

rare rare

unknown no

unknown na

rare rare

yes yes

1 day or more 15-18 days

rare

yes

15 days

rare rare

unknown yes

unknown < 2 weeks

In general, only case reports are available to assess these criteria. Dose response and specificity are aspects that could not be determined in this case. Bold-faced ans000000wers indicate criterion is positive for causation.

aspects of causation that will be addressed are type of study, magnitude of association, temporal sequence, consistency (from study to study), biological plausibility, and analogy (Table I).14,15 To begin this process, we performed searches for similar case reports (to address consistency), other studies of OPSI risk factors (addressing consistency and analogy), and studies of splenectomized animal models (for biological plausibility). Article searches. First, the published literature was searched to find any other reports of dental procedures, conditions, or oral bacteria associated with splenectomy. PUBMED was searched from 1965 to July 2003 by using the MeSH terms (1) “splenectomy” or (2) “splenic diseases” and “tooth”(textword) (n = 6) or (3) “periodontal”(textword) (n = 0). Additional search strategies used included “splenic diseases” and “streptococcus sanguis” (MeSH) (n = 1), along with other oral organisms such as S oralis, S mutans, S milleri, S sobrinis, S salivarius, Staphylococcus intermedius, Bacteroides gingivalis (n = 0), and Captocytophaga (n = 12). No cases of OPSI were located that were associated with either a dental extraction or another dental procedure. One report of a case that was not truly OPSI, but involved instead a liver abscess, showed a temporal association with a prior dental abscess.16 The tooth was extracted after symptoms of low-grade fever Journal of Evidence-Based Dental Practice Volume 3, Number 3

were present, and the liver infection was found and treated 10 days later. The bacteria cultured were S intermedius, an oral pathogen. Another case was reported in which OPSI was caused by S sanguis,17 a dental organism frequently associated with bacterial endocarditis.18 No obvious dental infections were present at time of admission, although it was not known whether a prior dental procedure had occurred (personal communication, 1994 Jul 12, H. Lars Aanning, MD). Second, the literature was searched for studies of the risk factors for infection in splenectomized patients. The PUBMED search for “splenectomy” [MeSH] and “infection” [MeSH] and “risk factors” [MeSH] yielded 67 articles. None of these articles evaluated dental or medical procedures, nor did they evaluate dental or other infections as risk factors for developing OPSI, bacteremia, or other complications after splenectomy. Of the few larger cohort studies in which multivariate analyses were used to adjust and control for differences between subjects, only the lack of prior pneumococcal vaccination was found to be a significant risk factor,3 although some others have found associations with immunosuppression and malignant neoplasms.19 Third, the literature was searched for articles that could address the question of biological plausibility. Drangsholt, Schubert, and Ritchie 159

TABLE 1I. Continued Biological plausibility Time sequence correct

Consistency with previous reports?

yes yes undetermined undetermined undetermined yes yes

no no yes yes no no no

undetermined yes

no yes

No. of case reports

2 >100

>1000

Animal models of splenectomy have been developed, and perhaps there were studies investigating oral organisms or dental procedures. A PUBMED search by using the terms “splenectomy” [MeSH] and “disease models, animal” [MeSH] yielded 152 articles, none of which were related to dental procedures or oral organisms. We also tried the terms “animal” [MeSH] and “splenectomy” [MeSH] and “periodontal” or “dental” or “tooth” but did not find any pertinent articles. Article and evidence evaluation. Since it is now generally accepted that one cannot prove causation, but instead must disprove competing hypotheses,20,21 we constructed multiple competing hypotheses of how the patient in this report developed the overwhelming postsplenectomy infection. These included the following: (1) a poorly healing and infected extraction site #2 in close proximity to the maxillary sinus created an S pneumoniae bacteremia; (2) a small oroantral fistula created after the surgical extraction #2 allowed S pneumoniae resident in the maxillary sinus to create an S pneumoniae bacteremia that caused the OPSI; (3) an occult maxillary sinus infection of S pneumoniae caused the OPSI; (4) a spontaneous nondental or sinus focus of infection of S pneumoniae caused the OPSI; (5) tonsillitis/oropharyngeal infection with S pneumoniae caused the OPSI; (6) periapical abscesses of #4, #19, and possibly #2 caused OPSI by means of an S pneumoniae bacteremia; (7) dental extractions #2, #4, #19 caused OPSI by means of an S pneumoniae bacteremia; (8) another uncultivated organism caused the OPSI. Obviously other hypotheses can be constructed, but these were thought to be some of the most likely or were brought up in discussions of this case. 160 Drangsholt, Schubert, and Ritchie

Animal models?

Plausible organism?

no no no no no no no

yes yes yes yes yes no no

no yes

undetermined yes

Analogy yes no no no no yes yes, hepatic abscess yes

We will now address each hypothesis according to each aspect of causation (Table II). First, only case reports, which are the generally acknowledged weakest level of evidence among human clinical studies, are available to address these causal hypotheses. Because there is no controlled evidence about dental risk factors, there are no odds ratios or relative risks to describe the magnitude of the association. The problem of assessing cause and effect for a single case report is similar to that of assessing cause and effect for adverse reactions to drugs or vaccines.22 In these cases, if the exposure is common, the outcome rare, and the association in time very close to the usual induction time, the probability of the association is higher.23 All the hypotheses have a rare outcome, which is OPSI; some the exposures are uncommon but occasionally occur, like pneumococcal oropharyngeal infection.24 oroantral fistula after tooth extraction,25 and occult maxillary sinusitis,26 with S pneumoniae as the infecting organism. Other exposures listed are either rare or have not been shown to occur. The usual induction period (the time from precipitating event to first symptoms) for OPSI is a few days to several weeks, although the upper limit of time is not well characterized. Although most of the hypotheses are within this time criterion, one appears not to—since the patient reported a sore throat only a day before he was admitted, and this may not have been enough time for a true tonsillitis with S pneumoniae to cause OPSI. In addition, published OPSI cases with a clear-cut event causing the infection, such as dog bites and Captonocytophaga infection, show induction periods from 2 to 8 days,27 casting some doubt on the dental extractions or periapical infections as likely hypothesis since they ocJournal of Evidence-Based Dental Practice September 2003

Fig 2. Possible causal chain of events in this case of overwhelming postsplenectomy infection. The bold text shows known events, and the regular text shows hypothesized events. The oroantral fistula pathway is 1 variation of a working hypothesis, while the leading hypothesis does not require this event. curred 15 to 18 days before the symptoms were first noted. A clear sequence of time with the exposure preceding the outcome is a necessary requirement for causation. Several of the hypotheses meet this criterion whereas, for several others, the current level of information does not allow us to determine whether the criterion is met. Are there other case reports describing a similar association between the purported exposure and outcome? The literature search showed only a handful, and none of these associated S pneumoniae OPSI with any type of dental procedure or dental infection. There are, however, case reports of occult maxillary sinus infection with S pneumoniae associated with OPSI,28,29 and many cases and case series of S pneumoniae OPSI without any known focus of infection.5,17,30 The biological plausibility of each hypothesis can be asJournal of Evidence-Based Dental Practice Volume 3, Number 3

sessed in two ways: by determining (1) whether there are existing animal models of splenectomy (eg, with splenectomized rats and induced dental infections) and (2) whether the organism has been plausibly associated with both the exposure and outcome. We could not locate any animal models simulating the hypothesized mechanisms, although over a thousand articles using various splenectomized animal models have been published. The type of bacterial organism—S pneumoniae—has been shown to be associated with all of the proposed hypotheses except for dental extractions and infections. S pneumoniae is very rarely found in the oral cavity, and has not been shown in the recent literature to be associated with these procedures.31-33 However, there is a growing body of evidence showing that maxillary tooth periapical abscesses are associated with maxillary sinusitis and that they are both infected with similar organisms.34 Drangsholt, Schubert, and Ritchie 161

Analogy is a weak causal criterion, and is dismissed by many. Nevertheless, as previously noted, there is a case report of S intermedius hepatic abscess developing after a prior odontogenic infection and dental extraction in an asplenic adult male that is analogous to the present case, the only difference being the specific type of systemic infection. Dose response and specificity are other causal criteria that were not seen as useful in this particular discussion, and are not addressed. Given the causal criteria addressed, the competing hypotheses are ranked in Table II from highest (1) to lowest (8) probability. A diagram of how two of the leading hypotheses could explain this phenomenon is depicted in Figure 2. This ranking is a qualitative assessment, and giving points and adding up scores is not a good method to help determine causation when different criteria carry different indeterminable weights in the decision-making process. The last hypothesis, that there was some other uncultivable organism, is not a testable hypothesis, so it does not shed much additional light on the question at hand. Although an essentially infinite number of hypotheses could be constructed, if they are not detailed enough, they will be similarly untestable and unhelpful. Comparing this case with other known dental procedure–systemic associations, such as dental extractions and bacterial endocarditis, shows the great difference in the level of evidence for OPSI and for dental-induced bacterial endocarditis (Table II). With bacterial endocarditis, there have been over a 1000 case reports published during the last 80 years that show a close temporal association; there are specific animal models of tooth extraction resulting in endocarditis; and there is controlled clinical evidence showing an increased risk from dental extractions in 3 of 4 of the published studies.4

Evidence summarization To summarize, no controlled clinical evidence currently exists assessing the relationship between OPSI and dental procedures or conditions. Only this current case report and a handful of others associate dental or oral mechanisms with OPSI, such as dental extraction or preceding odontogenic or postoperative infections. Currently accepted criteria for causation are discussed, with 8 competing clinical explanations of how the patient developed OPSI in this case. The lack of prior evidence of S pneumoniae found in the blood after dental procedures decreases the chance that the dental extraction was the main cause of the OPSI, although the extraction may have been related through alternative mechanisms as discussed previously. Thus, although this report is based on a single case, we believe the evi162 Drangsholt, Schubert, and Ritchie

dence shows it is more likely than not that the postoperative dental infection after the dental extractions was a contributing cause in the complex cascade of events leading to the OPSI. The level of evidence is undoubtedly weak, probably between level 4 to 5, since the conclusions are based on an evaluation of case reports. Nevertheless, in clinical practice, decisions and conclusions are often required to be made on weak evidence. We have attempted to show how a systematic evaluation of a case report by using evidence-based methodology can more convincingly address cause and effect in a single clinical case. More should be done to study the frequency, diagnosis, etiology, and treatment of postsplenectomy sepsis associated with precipitating dental or medical procedures and diseases.

Preventing OPSI in dental practice Assuming that the association between the dental interventions and overwhelming sepsis is one of cause and effect, what strategies can be recommended to prevent OPSI in dental practice (Table III)? Addressing this issue may be the most important part of this report, since up-to-date dental textbooks on managing medically compromised patients are notably silent on this entire subject,35,36 and there is evidence that current OPSI preventive practices are not being followed.37,38 First, health histories inquiring about past surgical operations and specifically about the history of splenectomy should be elicited from each patient. Patients with sickle cell disease, systemic lupus erythematosis, or Fanconi syndrome, in addition to a long list of other conditions, are also functionally asplenic and should be identified and educated about their increased risk.1(p.501) Second, the current 23-valent pneumococcal vaccine (Pneumovax 23, Merck Co, Whitehouse Station, NJ), recommended by several medical and public health organizations for asplenic patients,39,40 should be administered. A recent cohort study showed that vaccination reduced the risk of late bacteremia by about 75%.3 Each asplenic patient above the age of 2 should receive this vaccine, especially before invasive dental procedures. Medi-alert bracelets have also been recommended by some groups.1,39 Third, after any procedure, patients should be told of the warning signs of systemic infection and should be carefully monitored, with postoperative phone calls or visits. Fourth, although the efficacy of antibiotics in preventing OPSI after a dental procedure is unknown, it may be valuable to prescribe ampicillin 500 mg with clavulanic acid for 10 to 14 days after major procedures such as dental extractions or periodontal surgery.39 One animal model study has shown some efficacy of antibiotic prophylaxis for preventing Journal of Evidence-Based Dental Practice September 2003

OPSI,41 although a few antibiotic prophylaxis failures have been documented.42 This case report illustrates that the greatest risk of postsplenectomy sepsis may be due to dental infections and not necessarily dental procedures. US and British evidence-based guidelines have been developed to help prevent pneumococcal and other infections in anatomically and functionally asplenic individuals.39,40

SUMMARY People without a spleen or with one that is inadequately functioning have been shown to have a substantially increased lifetime risk for OPSI, a serious and often fatal condition. We present the first case of OPSI after dental extractions and postoperative infection—a 31-year-old man who developed a life-threatening S pneumoniae infection 15 days after dental extractions and 8 days after a follow-up appointment for a poorly healing extraction socket. Although his medical treatment was appropriate and prompt, he subsequently developed gangrene of his extremities as a consequence of the infection; this complication required bilateral belowthe-knee amputations and amputation of all his digits. An evidence-based discussion of determining cause and effect when a dental procedure or condition is associated with a poor systemic outcome is presented. Criteria to be considered include study design, magnitude of the association, temporal sequence, consistency from study to study, biological plausibility, dose response, and analogy. In this discussion, since only case reports are available to evaluate this relationship, the criteria are modified for the study design. After evaluating alternative hypotheses using these criteria, we conclude that the most likely cause of the sepsis was the development of postsurgical extraction infection, occurring in close proximity to a maxillary sinus colonized with S pneumoniae. Such evidence-based criteria can be used for other case reports to evaluate whether a cause-and-effect relationship occurred. Finally, dental practice guidelines to prevent these severe infections in asplenic patients are listed. REFERENCES 1. Brigden ML. Detection, education and management of the asplenic or hyposplenic patient. Am Fam Physician 2001;63:499506, 508. 2. Hosea SW, Brown EJ, Hamburger MI, Frank MM. Opsonic requirements for intravascular clearance after splenectomy. N Engl J Med 1981;304:245-50. 3. Ejstrud P, Kristensen B, Hansen JB, Madsen KM, Schonheyder HC, Sorensen HT. Risk and patterns of bacteraemia after splenectomy: a population-based study. Scand J Infect Dis 2000;32:52125. 4. Drangsholt MT. A new causal model of dental diseases associated with endocarditis. Ann Periodontol 1998;3:184-96.

Journal of Evidence-Based Dental Practice Volume 3, Number 3

TABLE III. Preventing infection in asplenic patients in dental practice Identify people at risk with a written/oral health history: Spleen removed Spleen not adequately functioning –Sickle cell anemia –Systemic lupus erythematosis –Fanconi syndrome –Other conditions.* Be certain the person has had the pneumococcal vaccine before invasive procedures are performed. Carefully consider the risks and benefits of prophylactic antibiotics before dental procedures that cause bacteremia, and use them if appropriate. Be watchful of any signs or symptoms of systemic infection after invasive dental procedures, and inform the patient to be vigilant. Contact the patient’s physician if any questions arise. *Many other conditions are associated functional asplenia. 5. Bisharat N, Omari H, Lavi I, Raz R. Risk of infection and death among post-splenectomy patients. J Infect 2001;43:182-86. 6. Holdsworth RJ, Irving AD, Cuschieri A. Postsplenectomy sepsis and its mortality rate: actual versus perceived risks. Br J Surg 1991;78:1031-38. 7. Davidson RN, Wall RA. Prevention and management of infections in patients without a spleen. Clin Microbiol Infect 2001;7:657-60. 8. Cullingford GL, Watkins DN, Watts AD, Mallon DF. Severe late postsplenectomy infection. Br J Surg 1991;78:716-21. 9. Lynch AM, Kapila R. Overwhelming postsplenectomy infection. Infect Dis Clin North Am 1996;10:693-707. 10. Terezhalmy GT, Hall EH. The asplenic patient: a consideration for antimicrobial prophylaxis. Oral Surg Oral Med Oral Pathol 1984;57:114-17. 11. Sarangi J, Coleby M, Trivella M, Reilly S. Prevention of post splenectomy sepsis: a population based approach. J Public Health Med 1997;19:208-12. 12. Sackett DL, Straus S, Richardson WS, Rosenberg W, Haynes RB. Asking answerable clinical questions. In: Evidence-based medicine. 2nd ed. Edinburgh: Churchill-Livingstone; 2000. p. 1328. 13. Sackett DL, Straus S, Richardson WS, Rosenberg W, Haynes RB. Therapy. In: Evidence-based medicine. 2nd ed. Edinburgh: Churchill Livingstone; 2000. p. 105-54. 14. Rothman K, Greenland SJ. Causation and Causal Inference. In: Rothman K, Greenland SJ, editors. Modern Epidemiology. 2nd ed. Philadelphia: Lippincott-Raven; 1998. p. 8-28. 15. Hill AB. The environment and disease: association or causation? Proc R Soc Med 1965;58:295-300. 16. Llibre JM, Cucurull J, Aloy A, Hernandez JA. Antimicrobial prophylaxis for dental extractions after splenectomy. Lancet 1991;337:1485-86. 17. White BP, Aanning HL. Overwhelming postsplenectomy sepsis twenty-two years after operation risks management and prevention. S D J Med 1991;44:317-20. 18. Banks J, Poole S, Nair SP, Lewthwaite J, Tabona P, McNab R, et al. Streptococcus sanquis secretes CD14-binding proteins that stimulate cytokine synthesis: a clue to the pathogenesis for infective (bacterial) endocarditis? Microb Pathog 2002;32:105-16.

Drangsholt, Schubert, and Ritchie 163

19. Schwartz PE, Sterioff S, Mucha P, Melton LJ III, Offord KP. Postsplenectomy sepsis and mortality in adults. JAMA 1982;248:227983. 20. Buck C. Popper’s philosophy for epidemiologists. Int J Epidemiol 1975;4:159-68. 21. Maclure M. Multivariate refutation of aetiological hypotheses in non-experimental epidemiology. Int J Epidemiol 1990;19:782-87. 22. Halsey NA. Anthrax vaccine and causlity assessment from individual case reports. Pharmacoepidemiol Drug Saf 2002;11:615-6. 23. Collet JP, MacDonald N, Cashman N, Pless R. Monitoring signals for vaccine safety: the assessment of individual adverse event reports by an expert advisory committee. Advisory Committee on Causality Assessment. Bull World Health Organ 2000;78:178-85. 24. Sass W, Bergholz M, Kehl A, Seifert J, Hamelmann H. Overwhelming infection after splenectomy in spite of some spleen remaining and splenosis. A case report. Klin Wochenschr 1983; 61:1075-79. 25. Hirata Y, Kino K, Nagaoka S, Miyamoto R, Yoshimasu H, Amagasa T. A clinical investigation of oro-maxillary sinus-perforation due to tooth extraction. Kokubyo Gakkai Zasshi 2001;68:249-53. 26. Penttila M, Savolainen S, Kiukaanniemi H, Forsblom B, Jousimies-Somer H. Bacterial findings in acute maxillary sinusitis—European study. Acta Otolaryngol Suppl 1997;529:165-68. 27. Pers C, Gahrn-Hansen B, Fredriksen W. Capnocytophaga canimorsus septicemia in Denmark, 1982-1995: review of 39 cases. Clin Infect Dis 1996;23:71-5. 28. Shih KY, Chu TS, Hung CC, Wu MS. Rhabdomyolysis associated with Streptococcus pneumoniae bacteremia in a splenectomized patient. J Formos Med Assoc 2002;101:429-31. 29. Ziske CG, Muller T. Partial splenectomy: uses of error. Lancet 30. Aksnes J, Abdelnoor M, Mathisen O. Risk factors associated with mortality and morbidity after elective splenectomy. Eur J Surg 1995;161:253-58. 31. Everett ED, Hirschmann JV. Transient bacteremia and endocarditis prophylaxis—review. Medicine 1977;56:61-77. 32. Hall G, Hedstrom SA, Heimdahl A, Nord CE. Prophylactic administration of penicillins for endocarditis does not reduce the in-

164 Drangsholt, Schubert, and Ritchie

33.

34. 35. 36. 37. 38. 39. 40.

41.

42.

cidence of postextraction bacteremia. Clin Infect Dis 1993;17:18894. Heimdahl A, Hall G, Hedberg M, Sandberg H, Soder PO, Tuner K, et al. Detection and quantitation by lysis-filtration of bacteremia after different oral surgical procedures. J Clin Microbiol 1990;28:2205-9. Brook I, Frazier EH, Gher ME. Microbiology of periapical abscesses and associated maxillary sinusitis. J Periodontol 1996;67: 608-10. Little JW, Falace DA, Miller CS, Rhodus NL. Dental management of the medically compromised patient. 6th ed. St. Louis: Mosby; 2002. Silverman S, Eversole LR, Truelove EL. Essentials of oral medicine. Hamilton (Ontario): BC Decker Inc;2001. White KS, Covington D, Churchill P, Maxwell JG, Norman KS, Clancy TV. Patient awareness of health precautions after splenectomy. Am J Infect Control 1991;19:36-41. Waghorn DJ. Overwhelming infection in asplenic patients: current best practice preventive measures are not being followed. J Clin Pathol 2001;54:214-18. Davies JM, Barnes R, Milligan D. Update of guidelines for the prevention and treatment of infection in patients with an absent or dysfunctional spleen. Clin Med 2002;2:440-3. Overturf GD. American Academy of Pediatrics. Committee on Infectious Diseases. Technical report: prevention of pneumococcal infections, including the use of pneumococcal conjugate and polysaccharide vaccines and antibiotic prophylaxis. Pediatrics 2000;106:367-76. Dickerman JD, Bolton E, Coil JA, Chalmer BJ, Jakab GJ. Protective effect of prophylactic penicillin on splenectomized mice exposed to an aerosolized suspension of type III Streptococcus pneumoniae. Blood 1979;53:498-503. Zarrabi MH, Rosner F. Rarity of failure of penicillin prophylaxis to prevent postsplenectomy sepsis. Arch Intern Med 1986;146: 1207-8.

Journal of Evidence-Based Dental Practice September 2003