Original Article
Hyperendemic Streptococcus pyogenes Infection despite Prophylaxis with Penicllin G Benzathine
N Engl J Med 1991; 325:92-97July 11, 1991
- Abstract
- Abstract
Background.
In closely confined populations, in which epidemics of Streptococcus pyogenes infection are common, penicillin G benzathine has long been used prophylactically to reduce morbidity from this pathogen. We report on our investigations of the effectiveness of penicillin G benzathine prophylaxis at a military recruit camp.
Methods.
We prospectively studied the rates of pharyngeal colonization and infection by S pyogenes among 736 male U.S. Marine Corps recruits from January througg MMrch 1989. Throat swwbs for culture, clinical data, ann questionnaire ddta were obttined during six examinations at intervvls of two weeks. Serum samples obtained bbfore training, after training, ann from acuttly ill recruits were analyzed with use of an antistrepttlysin O miirotttration technique.
Results.
Although 93 perrent of the recruits received prophylaxis with two intramuucular injections of 1.2 million units of penicillin G benzathine each (administered 30 to 39 days apart), 33 percent of the recruits were colonized by SS pyogenes, and 42 percent had infection (as ddfined byya two-dilution increase in the anttstreptolysin OOtiter). Thirty-seven percent of 265 recruits who reported a sore throat and were infected with S. pyogenes did not seek medical attention. The recruits who were allergic to penicillin (7 percent of the total), who received no prophylaxis, were more likely to be colonized; an increased risk of colonization and infection among the nonallergic recruits was associated with the presence of a higher percentage of allergic recruits in the platoon. After the study was completed, all recruits who were allergic to penicillin were prescribed 250 mg of oral erythromycin twice daily (a total daily dose of 500 mg) for 60 days. Subsequently, the average weekly rate of clini0cally evident S. pyogenes pharyngitis fell by more than 75 percent.
Conclusions.
If the prevention of S. pyogenes infection is to be effective in closely confined populations such as military recruits, prophylactic antibiotics must be administered to all members of the population. Exempting those who are allergic to penicillin may create a bacterial reservoir from which infection can be transmitted to nonallergic members of the population. (N Engl J Med 1991; 325:92–7.)
Media in This Article
Figure 1
Percentage of Throat Cultures Positive for S. pyogenes at Six Examinations, According to Penicillin Prophylaxis Schedule and Day of Training.Figure 2
Weekly Rates of S. pyogenes Pharyngitis during January through May 1988, 1989, and 1990.Article Activity
- Article
RECENT outbreaks of acute rheumatic fever1 2 3 and b. increases in the frequency of severe streptococcal infections4 5 6 7 8 are reminders of the severe morbidity that Streptococcus pyogenes may produce. Many of the S. pyogenes isolates described in recent reports have had a mucoid-colony morphology1 , 3 , 9 10 11 similar to that associated with severe streptococcal disease 20 years ago.12 , 13 These recent epidemics may reflect the reappearance of more virulent strains of S. pyogenes.9 , 12 13 14
Military recruits, who often share crowded living spaces, are at particularly high risk for streptococcal infection.15 , 16 During World War II, there were approximately 1 million streptococcal infections in the U.S. Navy17 and more than 21,000 cases of acute rheumatic fever.18 In 1944, rheumatic fever ranked second to fractures as a cause of time lost due to sickness or non-combat injury.18 The majority of cases occurred at military recruit camps. For 30 years, penicillin G benzathine, because of its ease of administration and assured compliance, has remained the drug of choice for preventing S. pyogenes infection in such high-risk populations.17 , 19 20 21 22 23 24 25 26 27 In response to reports that penicillin G benzathine was failing to prevent high rates of S. pyogenes pharyngitis at a recruit camp, we undertook to assess the effectiveness of this agent in preventing colonization and infection with S. pyogenes among U.S. Marine Corps recruits.
Methods
Study Group
Twelve platoons of male Marine Corps recruits who arrived for training at Marine Corps Recruit Depot, San Diego, California, during January 1989 were invited to participate in the study. Of the eligible recruits, 873 (99 percent) gave written informed consent. The 12 platoons (each made up of approximately 70 men) were divided into two companies that began 11 weeks of training one week apart and graduated in March 1989. All 74 drill instructors who directed the recruits' training participated in the study. Each drill instructor worked an estimated 90 to 95 hours per week with his platoon.
The recruits spent most of their training period in close contact with the other members of the platoon and with drill instructors. They were housed in open-bay barracks, except for one week when they slept in two-man tents. The recruits were not permitted outside the training areas until after graduation. Most meals were served in large dining facilities where the recruits were seated according to platoon. During one week of training, the recruits left the platoon structure and worked in dining facilities, as messengers, or as administrative assistants in other military camp locations.
Clinical Sampling
The participating recruits were enrolled in the study within 48 hours of arriving at the camp. A questionnaire was administered, a throat swab was obtained for culture, and a brief clinical examination was performed at the time the recruits were enrolled and again at intervals of approximately two weeks for a total of six examinations. Serum samples were collected during the first examination and 11 weeks later. Throat swabs were obtained for culture from the drill instructors at each examination, but the instructors did not complete questionnaires or undergo clinical examination.
The clinical examinations included a visual ins0pection of the posterior pharynx and tonsils and palpation for anterior cervical adenopathy. A recruit's pharynx was considered inflamed if erythema was present, with or without exudate. Adenopathy was defined as palpable tonsillar or submaxillary lymph nodes, with or without tenderness. The study did not interfere with routine medical care, which was readily available to the recruits.
Laboratory Studies
Throat swabs were plated onto 5 percent sheep-blood—agar plates within 24 hours of collection. The blood-agar plates were incubated aerobically at 35°C and examined after 24 and 48 hours. Beta-hemolytic colonies were screened for streptococcal group A antigen with the Reveal Colour Strep A latex system (Wellcome Diagnostics, Dartford, England). The minimal inhibitory concentration of penicillin G in S. pyogenes isolates was determined with use of standard techniques.28
Serum samples were tested for group A streptococcal antibodies by antistreptolysin O microtitration.29 , 30 Serum dilutions for the antistreptolysin O test were made with use of a 0.10 logarithmic scale, as follows: 1:100, 1:120, 1:160, 1:200, 1:240, 1:320, 1:400, 1:480, 1:640, 1:800, 1:960, 1:1280, 1:1600, 1:1920, and 1:2560. Bacto Streptolysin O reagent (Difco, Detroit) and dried Bacto Streptolysin O buffer (Difco) were prepared as described by the manufacturer. The Bacto Antistreptolysin O standard (Difco) was diluted to 1:25 in order to treat it in the same manner as the unknown serum. serologic tests were performed without knowledge of the throat-culture results. The results of serologic tests and serial throat cultures were not reported to the study investigators performing the clinical examinations or to the medical clinics that provided care to the recruits.
Data on Medical Care
All study participants who reported to a medical treatment facility with signs and symptoms of pharyngitis completed a questionnaire a throat culture was also prepared and a serum sample obtained. Any throat culture that was read as positive for S. pyogenes at a local medical treatment facility was inoculated into transport medium and sent to the study laboratory for confirmation. Computerized pharmacy data bases were used to obtain information on prescriptions for study participants.
Penicillin G Benzathine Prophylaxis
At the time of the study, the Navy's program to prevent streptococcal infection required the monitoring of weekly rates of S. pyogenes pharyngitis among recruits as a guide for the use of penicillin G benzathine.17 These rates were calculated from the number of recruits who sought medical attention and had confirmed (throat culture—positive) S. pyogenes pharyngitis. The routine scheduling of the penicillin G benzathine injections varied at each camp. Scheduling was based on weekly rates of S. pyogenes pharyngitis, studies of recruit populations conducted beginning more than 30 years ago,17 , 27 the camp's history of morbidity from S. pyogenes, logistical concerns, and a desire to provide maximal coverage for the duration of training with the fewest possible injections.
During calendar year 1988, all recruits were questioned about their history of penicillin allergy. Those who did not report an allergy were given 1.2 million units of intramuscular penicillin G benzathine on the 17th day of training and usually again on about the 55th day. According to Navy policy, recruits who reported a penicillin allergy, even if it was undocumented, did not receive prophylactic antibiotics. To allow us to determine if a shorter interval between injections would reduce the rate of S. pyogenes infection, 6 of the 12 platoons included in the study received the second injection on about the 55th day of training (routine schedule) and the other 6 platoons received the second injection on the 48th day of training (early schedule).
Definitions of Colonization and Infection
A recruit was considered to have been colonized when the results of serial throat cultures were first negative and later positive for S. pyogenes. A recruit who entered training with a positive throat culture and later had a negative culture was defined as having been colonized if a third or subsequent throat culture was positive for S. pyogenes. Infection was defined as a two-dilution increase in the antistreptolysin O titer (in comparison with the titer at the beginning of training), regardless of the results of culture. If more than two serum samples were obtained from a recruit (because he had an acute illness), the maximal antistreptolysin O value was compared with the titer before training.
Risk Factors
Possible risk factors for colonization and infection included age, the receipt of penicillin G benzathine, having an initially positive throat culture for S. pyogenes, the percentage of platoon members with positive S. pyogenes throat cultures at enrollment, having a drill instructor from whose throat culture S. pyogenes was isolated, the percentage of recruits allergic to penicillin within a platoon, the schedule for the administration of penicillin G benzathine (early vs. routine), race or ethnic group, company, the presence of a sore throat lasting more than six hours, feeling more tired than usual for reasons other than lack of sleep or excessive exercise, a cough that interfered with sleep, working in a dining facility, a history of a tonsillectomy before training, sharing a canteen with another recruit, and the presence of joint pain that exceeded that expected from heavy exertion. Clinical risk factors included an inflamed pharynx, cervical adenopathy, and the receipt of a course of 10 days or more of oral erythromycin. Erythromycin was prescribed by regular medical providers (not the investigators) for various conditions such as acne vulgaris.
Statistical Analysis
Questionnaire and laboratory data were merged and analyzed with use of SAS software (version 5.16, Cary, N.C.). Confidence limits for relative-risk calculations were determined by the method of Katz et al.31 , 32 Potential risk factors for colonization and infection were evaluated with the chi-square test. Biologic plausibility and an unadjusted P value of 0.25 were used as criteria for entering risk factors into the early multivariate modeks.33 Multivariate logisticregression modeling was performed with BMDP software (version 1987, University of California, Los Angeles). Regression coefficients were derived from unconditional maximum-likelihood ratios. Covariates and variable strata for the final multivariate models were selected after considering biologic relations and performing stepwise backward-elimination multivariate modeling.33 P value limits of more than 0.15 (to remove a variable) and less than 0.10 (to enter a variable) were used in the final multivariate models.
Results
The recruits enrolled in this study came from 38 states, Guam, and Puerto Rico. The racial and ethnic distribution of the 873 recruits enrolled was as follows: non-Hispanic white, 591 (67.7 percent); black, 95 (10.9 percent); Hispanic, 93 (10.7 percent); Filipino, 19 (2.2 percent); and other, 75 (8.6 percent). The median age of the recruits was 19 years (range, 17 to 29). The median duration of observation of individual recruits was 79 days (range, 1 to 80). Seven hundred fifty-nine of 812 recruits (93.5 percent) still in the study received the first scheduled injection of penicillin G benzathine, and 705 of those 759 (92.9 percent) received the second scheduled injection. Of the 873 recruits enrolled in the study, 736 (84.3 percent) were followed up for 75 days or more and provided two or more samples of serum. These recruits were studied for risk-factor associations.
Isolates of S. pyogenes
The proportion of recruits' throat cultures that were positive for S. pyogenes varied according to prophylaxis group, but overall it increased with time from 5.6 percent of 861 cultures initially to 12.3 percent of 741 cultures at the final examination (Fig. 1Figure 1
Percentage of Throat Cultures Positive for S. pyogenes at Six Examinations, According to Penicillin Prophylaxis Schedule and Day of Training.). thirty-six (7.3 percent) of the 491 S. pyogenes isolates collected in this investigation were mucoid. Only 10 of 340 isolates tested (2.9 percent) demonstrated minimal inhibition in response to a penicillin G concentration ≥0.025 μg per milliliter.Colonization and Infection
Of the 736 recruits who were followed for at least 75 days, 241 (32.7 percent) were identified as having been colonized by S. pyogenes sometime during training. Thirty-one percent of the recruits who received penicillin G benzathine were colonized, as compared with 54.2 percent of those who received no prophylaxis. The unadjusted relative risk of colonization among the recruits who received penicillin G benzathine, as compared with those who did not, was 0.6 (95 percent confidence interval, 0.4 to 0.8) (Table 1Table 1
Unadjusted Relative Risk of S. pyogenes Colonization or Infection of the Pharynx among Recruits with 75 or More Days of Exposure.*).The median antistreptolysin O titers for the 719 recruits from whom serum was obtained at both the first and the sixth examination were 1:200 before training and 1:320 after training. By substituting the antistreptolysin O values measured in serum obtained when a recruit had acute pharyngitis for missing posttraining serum samples, we were able to include 17 more recruits in the analysis of the outcome of infection. Three hundred six (41.6 percent) of the 736 recruits whom we followed for 75 days became infected. There was no difference in infection rates between those who received penicillin G benzathine and those who did not (Table 1). Forty-one (13.4 percent) of those who had infection did not report having a sore throat lasting for six or more hours. Ninety-seven (36.6 percent) of the 265 recruits who had infection and reported a sore throat did not seek medical attention. Altogether, 128 (41.8 percent) of the recruits with infections did not report to medical clinics. No recruit had acute rheumatic fever, streptococcal toxic shock—like syndrome, or acute glomerulonephritis during the study period.
Risk Factors
The following unadjusted risk factors were associated with a higher risk of colonization: a higher percentage of penicillin-allergic recruits in the platoon, the presence of cervical adenopathy, feeling unusually tired, the presence of sore throat, a higher percentage of culture-positive recruits in the platoon at enrollment, and the presence of cough. Receiving penicillin G benzathine was associated with a lower risk of colonization, and the risk of colonization also varied according to the recruit's company and age.
The following unadjusted risk factors were associated with a higher risk of infection: a higher percentage of penicillin-allergic recruits in the platoon, the presence of cervical adenopathy, receiving the second injection of penicillin G benzathine on the early schedule, feeling unusually tired, having been colonized, and having a cough; the risk of infection also varied depending on whether the drill instructor had had a throat culture positive for S. pyogenes. Sharing canteens, working in a dining facility, a history of tonsillectomy, and race or ethnic group were not statistically significant risk factors. The final multivariate model for colonization included only the following covariates as factors increasing the risk of colonization: a higher percentage of allergic recruits in the platoon, the presence of cervical adenopathy, feeling unusually tired, the presence of sore throat, and membership in Company 1 (Table 2Table 2
Adjusted Relative Odds of S. pyogenes Colonization or Infection of the Pharynx among Recruits with 75 or More Days of Exposure.*); receiving penicillin G benzathine reduced the risk of colonization. The final multivariate model for infection included only the percentage of allergic recruits in the platoon, the presence of cervical adenopathy, receiving erythromycin, having an initially positive throat culture, having been colonized, and age (Table 2).Changes in the Preventive Program after the Study
In October 1989, the training schedule for recruits was altered to reduce the interval between the first and second injections of penicillin G benzathine from 38 or 39 days to 30 days. Beginning in December 1989, after examining data from this investigation and consulting experts, the Navy decided to prescribe low-dose oral erythromycin base (250 mg twice a day for 30 days) for all penicillin-allergic recruits whenever penicillin G benzathine was administered. After these changes were made, weekly rates of S. pyogenes pharyngitis at the camp declined more than 75 percent as compared with the rates for the previous two years (Fig. 2Figure 2
Weekly Rates of S. pyogenes Pharyngitis during January through May 1988, 1989, and 1990.).Discussion
The high rates of S. pyogenes pharyngitis at this recruit camp and the high proportion of S. pyogenes infections observed in this study probably reflect a hyperendemic state of S. pyogenes infection.34 This state was probably due in part to the presence in the camp of the 7 percent of recruits who reported penicillin allergy and therefore received no prophylaxis. This proportion was higher than the 1 to 2 percent reported more than 30 years ago.17 Recruits who reported an allergy to penicillin were more likely to be colonized by S. pyogenes than nonallergic recruits and equally likely to be infected (Table 1). High proportions of penicillin-allergic recruits had throat cultures that were positive for S. pyogenes when the prevalence of S. pyogenes in other recruits had been nearly eliminated by the first injection of penicillin G benzathine (Fig. 1). Even when adjustments were made for other important risk factors (Table 2), belonging to a platoon with a high proportion of penicillin-allergic recruits increased both the risk of colonization (adjusted relative odds, 1.7) and the risk of infection (adjusted relative odds, 1.4). Thus, penicillin-allergic recruits probably served as a reservoir for the bacteria, providing a constant opportunity for S. pyogenes to colonize and infect other recruits.
Scheduling penicillin G benzathine injections 38 to 39 days apart probably also contributed to the high rates of streptococcal infection. Those who received the second injection earlier (after 31 days) had a higher proportion of cultures positive for S. pyogenes (12.8 percent) 15 days after the first injection than the recruits assigned to the routine schedule (6.0 percent); by the final sampling period, however, fewer recruits assigned to the early schedule had positive cultures (9.7 percent vs. 14.2 percent), suggesting that the earlier second injection of penicillin G benzathine was more effective (Fig. 1). It has been shown that when a similar regimen is followed, penicillin levels may decrease in effectiveness in as little as three weeks.35 , 36 Thus, it is very likely that some recruits were no longer protected by 38 days after the first injection. It is fortunate that despite the high infection rates, no rheumatic fever or toxic shock—like syndrome was observed. The absence of these diseases may have been due to a high prevalence of less virulent strains of S. pyogenes. Alternatively, the penicillin G benzathine may have been sufficient to prevent severe illness.
The results of multivariate modeling (Table 2) should be viewed as exploratory. As in previous studies, however, we found no association between a history of tonsillectomy,37 an inflamed pharynx,38 , 39 or a history of cough38 and the outcome of colonization or infection. Cervical adenopathy, found to be of mixed predictive value in previous reports.38 , 39 was associated with both colonization and infection in our study. A history of sore throat, identified as an important risk factor in previous studies,38 , 39 was important for colonization but not for infection among the recruits we studied.
Two adjusted risk factors, receiving a 10-day course of oral erythromycin and beginning training with a throat culture positive for S. pyogenes, were found to be protective against infection. The protective effect of erythromycin, often given at low doses, was similar to the effect of using this drug as an alternative form of prophylaxis (Fig. 2). The multivariate models showed that recruits who were carrying S. pyogenes when they began training were not at increased risk of infection, nor were the other members of their platoons. These results and the difficulty of eradicating the prolonged carrier state38 , 40 support the policy decision not to attempt to eradicate asymptomatic carriage of S. pyogenes when recruits first enter training.
The results of the multivariate analysis underscore the difficulty of predicting S. pyogenes infection solely on the basis of the clinical examination or symptoms. In addition, many infected subjects, like the 41.8 percent of infected recruits in this study, may never seek medical attention. Thus, a system of passive surveillance for clinical disease cannot avert streptococcal epidemics. Until we have a more effective technique to prevent rheumatic fever, such as a vaccine,41 , 42 prophylactic treatment of every member of susceptible populations is required for success.
This epidemiologic study has a number of limitations. Single throat cultures, which have a sensitivity of 90 to 95 percent41 , 43 under the best conditions, were prepared only at two-week intervals; as is obvious from the high infection rates, numerous colonizations were missed. No typing of S. pyogenes strains was performed; therefore, we could not distinguish between prolonged carriage of one strain and colonization with a new strain. Some of the infections may have occurred before prophylaxis was instituted. Because the administration of penicillin G benzathine may reduce antibody response44 , 45 and because only one serologic test was used, the true rates of infection may have been higher than those we found. Moreover, it is possible that the decline in the average weekly rates of pharyngitis in 1990 (Fig. 2) may have been caused by an as yet unidentified factor, such as cyclic Variation, unrelated to the interventions carried out after the end of our study.
In summary, this investigation has shown the importance of giving carefully scheduled prophylaxis against S. pyogenes to all members of high-risk populations. If portions of such populations are neglected, hyperendemic S. pyogenes infection may still occur. Our study also illustrates the difficulty of diagnosing S. pyogenes infection only on the basis of clinical signs and symptoms. Relying on a system of passive clinical surveillance to prevent S. pyogenes epidemics is inadequate for high-risk groups.
Supported by the U.S. Naval Medical Research Institute (work unit 3MI12770A870AR112).
Presented in part at the 39th annual meeting of the American Society of Tropical Medicine and Hygiene, November 1190,,New Orllans.
The opinions or assertions contained herein are the private views of the authors and are not to be construed as reflecting the official views of the U.S. Navy or the Department of Defense.
We are indebted to the following for their assistance in conducting the investigation: Capt. George J. Hansel, M.S.C., U.S.N., Col. Michael C. Bell, U.S.M.C., Capt. Harold G. McNeil, M.C., U.S.N., Capt. Robert W. Browning, M.C., U.S.N., Col. Richard N. Miller, M.C., U.S.A.., Capt. Fred L. Baker, M.C., U.S.N., Capt. Edward C. Oldfield, M.C., U.S.N., Comdr. Robert C. Rockhill, M.S.C., U.S.N., Lt. Comdr. Konrad E. Hayashi, M.C., U.S.N., Lt. Comdr. Sharon K. Wallace, M.C., U.S.N., Lt. Comdr. Bruce Merrell, M.S.C., U.S.N., Lt. Comdr. Daniel Angelier, M.S.C., U.S.N., Dr. Frank J. Goldzer, Maj. John G. McNeil, M.C., U.S.A.., Lt. Comdr. Douglas D. Slaten, M.C., U.S.N., Lt. Comdr. Gary L. Hubbard, M.S.C., U.S.N.R., Lt. Floyd A. Doughty, M.C., U.S.N., Lt. Judy A. McCloskey, N.C., U.S.N., Lt. Steven M.Jaeger, M.C., U.S.N.R., and Lt. James E. Roane, M.S.C., U.S.N.; to Mr. Charles V. Gray for his computer programming; to Capt. S. William Berg, M.C., U.S.N., for his review of the manuscript; and to the following dedicated Navy hospital corpsmen for their field and laboratory work: Mark A. Scales, HMC, U.S.N., Raymond W. Angel, HMC, U.S.N., Robert Martini, HMC, U.S.N., Chelton E. Jenkins, HMC, U.S.N., Marsha A. Roberts, HM1, U.S.N., Steve L. Martin, HM1, U.S.N., Regina A. Collantes, HM1, U.S.N., Elizabeth A. Richards, HM1, U.S.N., Boccaccio B. Aying, HM1, U.S.N., Mike W. Wilson, HM1, U.S.N., Mary L. Welborn, HM1, U.S.N., Peter J. Franke, HM1, U.S.N., James E. Bare, HM1, U.S.N.R., Richard P. Grant, HM1, U.S.N., Edwin A. Cordero, HM1, U.S.N., Joseph P. Goularte, HM2, U.S.N., Igmedio M. Osalla, HM2, U.S.N., Joseph J. Foust, HM2, U.S.N., Kristi K. Jenson, HM3, U.S.N., James B. Shaw, HM2, U.S.N., Brian D. Meier, HM3, U.S.N., and Harry T. Johnson, HM3, U.S.N.
Source Information
From the Epidemiology Department (G.C.G., J.P.S.) and the Microbiology Department (J.E., A.K.T.), Navy Environmental and Preventive Medicine Unit No. 5, San Diego, Calif.; the Epidemiology Division of the Naval Medical Research Institute, Bethesda, Md. (K.C.H.); and the World Health Organization Collaborating Center for Reference and Research on Streptococci, Department of Pediatrics, University of Minnesota, Minneapolis (E.L.K.). Address reprint requests to Commander Gray at the U.S. Naval Medical Rssarcc Unit No. 33 FFP NY 09527–1600.
- References
References
1
Veasy
Full Text | Web of Science | Medline2
Wallace
CrossRef | Web of Science | Medline3
Acute rheumatic fever among Army trainees — Fort Leonard Wood, Missouri, 1987–1988 . MMWR 1988; 37:519–22.
Medline4
Cone
Full Text | Web of Science | Medline5
Bartter
CrossRef | Web of Science | Medline6
Stevens
Full Text | Web of Science | Medline7
Thomas
CrossRef | Web of Science | Medline8
Group A beta-hemolytic streptococcal bacteremia — Colorado, 1989 . MMWR 1990; 39:3–11.
Medline9
Marcon
Web of Science | Medline10
Kaplan
CrossRef | Web of Science | Medline11
James
Full Text | Web of Science | Medline12
Stollerman
CrossRef | Web of Science | MedlineErratum, Arch Intern Med 1988; 148:2600.
Web of Science13
Bisno
CrossRef | Web of Science | Medline14
Schwartz
CrossRef | Web of Science | Medline15
Group A beta-hemolytic streptococcal pharyngitis among U.S. Air Force trainees — Texas, 1988–89 . MMWR 1990; 39:11–3.
Medline16
Jarrett
CrossRef | Web of Science | Medline17
Thomas
CrossRef | Medline18
History and accomplishments: an introduction to NAMRU-4. Great Lakes, 111.: Naval Medical Research Unit No. 4, 1972.
19
Brooks
Web of Science | Medline20
Frank
Medline21
Morris
Web of Science | Medline22
Davis
Full Text | Web of Science | Medline23
Schreier
Full Text | Medline24
Markowitz
Web of Science | Medline25
Dajani
Web of Science | Medline26
Rheumatic fever and rheumatic heart disease: report of a WHO study group . WHO Tech Rep Ser 1988; 764:1–58.
Medline27
Frank
Web of Science | Medline28
Washington JA. Antimicrobial susceptibility tests of bacteria. In: Washington JA, ed. Laboratory procedures in clinical microbiology. Boston: Little, Brown, 1974:291–7.
29
Klein
Medline30
Klein
Web of Science | Medline31
Katz
CrossRef | Web of Science32
Kahn HA, Sempos CT. Statistical methods in epidemiology. New York: Oxford University Press, 1989.
33
Hosmer DW Jr, Lemeshow S. Applied logistic regression. New York: John Wiley, 1989:82–134.
34
Last JM. A dictionary of epidemiology. 2nded. New York: Oxford University Press, 1988.
35
Kaplan
CrossRef | Web of Science | Medline36
Ayoub
CrossRef | Web of Science | Medline37
Meyer
Web of Science | Medline38
Kaplan
CrossRef | Web of Science | Medline39
Siegel
Full Text | Web of Science40
Kaplan
CrossRef | Web of Science | Medline41
Bisno AL. Management of streptococcal infections and prevention of rheumatic fever: the changing scene. In: Sande MA, Hudson LD, Root RK, eds. Respiratory infections. Vol. 5 of Contemporary issues in infectious diseases. New York: Churchill Livingstone, 1986:217–34.
42
Fischetti
CrossRef | Web of Science | Medline43
Kellogg
Web of Science | Medline44
Brock
CrossRef | Web of Science | Medline45
Chamovitz
Full Text | Web of Science | Medline
- Citing Articles (9)
Citing Articles
1
Seung-eun Lee, Angelia Eick, Paul Ciminera. (2008) Respiratory Disease in Army Recruits. American Journal of Preventive Medicine 34:5, 389-395
CrossRef2
Vivek Kak. (2007) Infections in Confined Spaces: Cruise Ships, Military Barracks, and College Dormitories. Infectious Disease Clinics of North America 21:3, 773-784
CrossRef3
M. K. O'Shea, M. A. K. Ryan, A. W. Hawksworth, B. J. Alsip, G. C. Gray. (2005) Symptomatic Respiratory Syncytial Virus Infection in Previously Healthy Young Adults Living in a Crowded Military Environment. Clinical Infectious Diseases 41:3, 311-317
CrossRef4
S. D. Putnam, G. C. Gray, D. L. Biendenbach and, R. N. Jones. (2001) . Response. Clinical Microbiology and Infection 7:7, 397-398
CrossRef5
G. C. Gray, P. R. Goswami, M. D. Malasig, A. W. Hawksworth, D. H. Trump, M. A. Ryan, D. P. Schnurr. (2000) Adult Adenovirus Infections: Loss of Orphaned Vaccines Precipitates Military Respiratory Disease Epidemics. Clinical Infectious Diseases 31:3, 663-670
CrossRef6
S. D. Putnam, G. C. Gray, D. J. Biedenbach, R. N. Jones. (2000) Pharyngeal colonization prevalence rates for Streptococcus pyogenes and Streptococcus pneumoniae in a respiratory chemoprophylaxis intervention study using azithromycin. Clinical Microbiology and Infection 6:1, 2-8
CrossRef7
Jeremy D. Gradon. (1996) SPACE-OCCUPYING AND LIFE-THREATENING INFECTIONS OF THE HEAD, NECK, AND THORAX. Infectious Disease Clinics of North America 10:4, 857-878
CrossRef8
E.J. Ridgway, K.D. Allen. (1993) Clustering of Group A streptococcal infections on a burns unit: important lessons in outbreak management. Journal of Hospital Infection 25:3, 173-182
CrossRef9
Denny, Floyd W., . (1991) The Streptococcus Saga Continues. New England Journal of Medicine 325:2, 127-128
Full Text
