Pediatric Plague

ABSTRACT.The largereservoirof animalplaguein the American West led to 121 cases of human plague from 1970 to 1980. The majority (55%) of recent cases have occurred in children 16years of age and younger. In New Mexico 38 pediatric plague cases were reviewed to deter mine the epidemiologic, clinical, and laboratory features of this disease. Thirty-one patients (82%) had bubonic plague and seven (18%) had primary septicemic disease. Primary septicemic plague had a significantly higher case fatality ratio (71%vs 3%;P = .0002)and an increased risk of plague pneumonia (57%vs 6%;P = .01) compared with bubomc disease. Symptoms at onset, physical examina tion, and laboratory data at hospitalization (mean of three days after onset) were consistent with an acute, systemic febrile illness. Recovery from plague was slow, requiring an average of 5.9 days from initiation of effective anti biotics until fever lysis. Only a minority ofplague patients were initially suspectedto have plague, even by the time of hospitalization. Whereas clinical evidence, particularly in bubomc disease, should suggest plague, residing in or visiting a rural area of the West (especially from June through September) during the week prior to illness may be the only useful epidemiologic clue for the majority of patients who lack a history suggestive of exposure to

animals or fleas. The importance of pediatric plague stems from the recent increase in cases, the significant increase in the proportion of all cases occurring in chil then, the public health implications of plague pneumonia (16%of cases),and the demonstrated potential for plague patients to travel to areas of the country unfamiliar with the disease and its sylvatic home in the American West. Pediatrics 69:762—767,1982;plague, Yersinia pestis.

reported plague cases in this country occurred among children 16 years of age or younger. During this period, approximately two thirds of these cases were reported in New Mexico.2 The number of pediatric plague cases and the lack of a general description of pediatric plague in the literature prompted this review of 38 cases of plague in chil then in New Mexico.

MATERIALS AND METHODS

All physician-diagnosed cases of human plague in New Mexico were legally required to be reported to the New Mexico Health and Environment Depart ment (NMHED). Documentation was provided by the state public health laboratory where fluorescent antibody (FA) and phage susceptibility testing of suspected Yersinia pestis isolates were performed. In addition, the Centers for Disease Control (CDC) reported results of all their plague cultures or se rologic tests to NMHED.

From 1970 to 1980 40 plague cases were reported in children 16 years of age or younger. Epidemio logic, clinical, and laboratory information from hos pital and NMHED records was compiled for 38 of these pediatric cases. Data were unavailable from two 1970 patients, who are not included in this review.

Definitions

Confirmed cases of plague required either isola tion of Y pestis from a clinical specimen or a four fold or greater rise or fall in passive hemagglutina tion titer ofp!ague antibody in suitably timed serum specimens. All New Mexico cases included in this report were confirmed by the CDC.

Bubonic plague was defined as Ypestis infection with objective evidence of lymphadenitis. Primary septicemic plague was Y pestis infection without objective evidence of adenopathy. Meningeal The large reservoir of animal plague established

in the American West during this century1 ensures that plague will continue to occur in residents and travelers to this area. From 1950 to 1979, 55% of all

Received for publication Oct 26, 1981; accepted Dec 10, 1981. Reprint requests to (J.M.M.) Health Services Division, New Mexico Health and Environment Department, P0 Box 968, Santa Fe, NM 87503.

PEDIATRICS (ISSN 00314005).Copyright ©1982by the American Academy of Pediatrics,

Pediatric Plague

Jonathan M. Mann, MD, MPH, Laurence Shandler, MD, FAAP, and Alice H. Cushing, MD, FAAP

TABLE 1. Selected with Pediatric Plague,Clinical

Features of 38 Patients

New Mexico, 1970 to 1980*Clinical

FeatureNo. FeatureBubonicof Patients with

Septi- Total

n=31 cemic n=38

n=7Plague

pneumonia2 (6) 4 (57) 6

(16)tPlague

meningitis3 (10) 1 (14) 4

(11)Fatal

case1 (3) 5 (71) 6 (16)@

unilateral in 90% and bilateral in one case each of inguinal, cervical, and axillary bubonic plague. Seven patients (18%) had primary septicemic plague. There were no differences in plague type according to age, sex, or ethnic group. However,

primary septicemic patients were significantly more

likely to develop plague pneumonia and die than were bubomc patients (Table 1).

Patients with plague pneumonia waited an aver age of 6.2 days from onset of symptoms before

beginning effective antibiotic therapy or until death

for those patients who never received an appropri ate antibiotic. In contrast, 29 patients without pneu monia or meningitis first received an effective an tibiotic after an average of 3.3 days after onset. Delays of four days or longer from onset to appro priate antibiotic treatment were significantly more common among patients with plague pneumonia

(6/6 vs 9/29; Fisher's exact test, P = .003).

Plague meningitis patients each had a different clinical course. One patient died approximately 2.5 days after onset of primary septicemic plague; Y

pestis was obtained from a culture of cerebrospinal

fluid obtained shortly before death. Two patients

developed meningitis while receiving an effective antibiotic. Meningeal signs and symptoms occurred in one of these patients following seven days of intramuscular streptomycin therapy and three days of oral tetracycline. The other patient received low dose oral tetracycline (18 mg/kg/day) starting on the second day of illness and continuing for one week prior to the onset of meningitis. The fourth patient developed meningitis after eight days of oral

penicillin therapy (including three days with oral

cephalexin) which was started on the second day of ifiness.

Onset of Plague and Initial Medical Management Symptoms of plague reported for the onset of ifiness are shown in Table 2. The collective experi

ence of 38 patients showed that patients first sought medical attention after an average of 1.5 days of plague required isolation of Y pestis from cerebro

spinal fluid. Pneumonia in association with docu mented plague infection was classffied as either

confirmed or probable.3 Confirmed cases had a pos

itive sputum or endotracheal culture for Ypestis or postmortem confirmation of plague pneumonia, along with typical signs and symptoms of pneumo nia. A probable case of plague pneumonia required: (1) signs and symptoms of pneumonia; (2) a nega tive sputum culture or FA test, or else sputum for culture was not obtained prior to antibiotic therapy and no FA test was performed; and (3) no other

likely explanation for the pulmonary parenchymal process.

RESULTS

Epidemiology

At least one case of pediatric plague was reported in each year of the study period except during 1971 to 1973. Of 38 cases, 37 (97%) occurred from June through September; the remaining case was re ported in February. Twenty of the 38 (53%) were girls. The mean age of all patients was 8.8 years (median 8.5 years, range 2 to 16 years). Sixteen patients (42%) were Hispanic, 14 were non-Hispanic whites (37%), and eight were Native Americans (21%). When only the 12 counties actually reporting plague in New Mexico are considered, the incidence rate for pediatric plague from 1970 to 1980 was 1.3 cases/105 person-years. Incidence rates varied among ethnic groups in the 12-county area: the Native American rate of 2.1 cases/105 person-years

was 50% higher than the Hispanic rate (1.4 cases!

io@

person-years)

andwasmorethantwicethenon

Hispanic white incidence rate (0.96 cases/105 per son-years).

The overall case-fatality ratio was 15.8% (6 deaths among 38 patients), with no significant differences among ethnic groups, between the sexes, or accord ing to the patients' age.

Environmental investigators generally found ev

idence of plague epizootic activity in the vicinity of the patient's home or in the epidemiologically im plicated habitat. In 36 of 38 cases, plague-infected fleas presumably transmitted the disease; one case was linked to skinning a dead coyote and another case was cat associated.

Clinical Characteristics

Thirty-one patients (82%) had bubonic plague. Inguinal/femoral buboes were most common (21/

30; 70%); cervical (5/30; 17%) and axillary locations (4/30; 13%) accounted for the remainder (bubo lo cation was not specified in one case). Buboes were

* Terminology has been defined in text. Percent of pa

tients is shown in parentheses.

t 2/31vs4/7;Fisher's

exacttest,P = .01.

SymptomNo. SymptomBubonicSeptiof Patients with

Totaln

= 31cemic@n = 38n=7Fever30

(97)6 (86)36 (95)Node-associated

pain21 (68)021

(55)Vomiting16

(52)3 (43)19 (50)Lethargy,

malaise, or12 (39)3 (43)15 (39)anorexiaHeadache11

(35)011 (29)Chills10

(32)1 (14)t11 (29)Abdominal

distress or8 (26)2 (29)10 (26)nauseaDiarrhea3

(10)03 (8)Cough2

(6)@02 (5)

TABLE 2. SymptomsReportedat Onsetof Illnessin

38 Patients with PediatricPlague,NewMexico,1970to

Laboratory Data

A total leukocyte count and differential was ob

tamed on hospital admissionfor 33 patients. The

mean leukocyte count was 13,400/cu mm (median

of 11,700/cu mm; range 4,600 to 52,700/cu mm). Neutrophilia (>70% neutrophils) and/or a left shift

( 10%bandforms)waspresentin 29of33patients

(88%). No other laboratory measurements were

available from a sufficient number of patients to enable a composite description.

Diagnosis of Plague

The diagnosis of plague was confirmed by culture in 35 patients and by serologic test (passive hemag glutination) in three patients. The 31 bubonic pa tients had cultures taken from either bubo aspirate, or blood, or both. Of these patients 24 had cultures

taken from both bubo aspirates and blood; both cultures were positive in 12 patients; six had positive

node aspirates only; three had positive blood cul tures only; and in three patients neither culture grew Ypestis (serologically confirmed cases). Bubo aspirates were not obtained from four patients; blood cultures were positive for Y pestis in two patients and were negative for the other two pa tients (both confirmed with positive CSF cultures).

Finally, in three bubonic patients, a node aspirate

was confirmatory and blood cultures were not ob

tamed. Overall, bubo aspirates were positive for Y

pestis in 21 of 27 patients (78%) and blood cultures were positive for 17 of 28 bubonic patients (61%); the combination of bubo aspirates and/or blood cultures provided culture confirmation in 26 of 31 bubonic patients (84%).

A rapid presumptive diagnosis of plague was pro vided by FA staining of node aspirate material from 12 patients; the total number of aspirate specimens

examined by FA technique was not known. Finally,

blood cultures were positive for Ypestis in all seven

primary septicemic patients.

Antibiotic Therapy

Thirty-four patients received an antibiotic con sidered effective against Y pestis, including strep tomycin (25 patients), tetracycline (22 patients), or chloramphenicol (11 patients). One patient received

gentamicin. A single antibiotic was administered to

13 patients; 18 were given two drugs (streptomycin and tetracycline accounted for two thirds of these regimens); two patients received three drugs; and one patient received streptomycin, tetracycline, gentamicin, and chloramphenicol. Of the six pa

tients who died, four never received an effective

antibiotic; the other two had been given one or two appropriate antibiotics on the day of death only.

* Terminology has been defined in text. Percent of pa

tients is shown in parentheses.

t Singlesepticemic

patientwhodidnotreportfeverat

onset.

:j:Neither patient had any evidence of plague pneumonia.

illness (range: same day to six days). This period

was identical for bubonic and septicemic patients and did not differ significantly between survivors and those for whom the disease was fatal. In only three instances (8%) was plague considered at the first medical encounter. Patients with plague were hospitalized an average of 1.5 days after the first medical contact. Septicemic patients took some what, but not significantly, longer to be hospitalized (mean of 1.9 days after first medical contact vs 1.4 days for bubonic patients).

Four patients died within a few hours of hospi

talization. Plague was considered in the differential

diagnosis at hospitalization for 13 of the remaining 33 patients (39%). One patient, recognized as having uncomplicated bubonic plague, was managed as an outpatient. The average time from onset of ifiness to death for the six fatal cases was 4.1 days, with a range of 2.5 to seven days.

Physical Examination

Temperatures recorded at hospital admission for 36 patients ranged from 37 to 41.1 C with a mean and median of 39.5 C. Only two patients (6%) had an admission temperature of less than 38.7 C. Both described subjective fever prior to hospitalization and each subsequently had temperatures of greater than 38.7 C recorded during hospitalization.

MeanMedianRangeMarked

symptomatic3.431-8improvement@Fever

[email protected]—15Incision and drainage of19.41513—46bubollDuration

of hospitalization14141-30 Hospital Course

Several features of the clinical course of patients who survived pediatric plague are shown in Table 3. There was no association between the illness day on which the antibiotic was started and the time until symptomatic improvement occurred, nor was

there an association between the day of illness the

antibiotic was started, the specific antibiotic used, or administration of single vs multiple drugs and the time until fever lysis. Rapid symptomatic im provement was significantly associated with early fever lysis. Of patients with symptomatic improve ment in three days or less, 75% had fever lysis within five days; in contrast, for those patients

requiring longer than three days to improve symp tomatically, only 30% had fever lysis within five days (Fisher's exact test; P = .03).

Systematic follow-up of 11 cases occurring in 1975 found no evidence of functional disability one month after hospital discharge. In addition, no neu rologic sequelae were observed in three patients who survived plague meningitis one year after re covery.

DISCUSSION

This review of 38 cases of pediatric plague in New Mexico illustrates the essential features of a disease which, due to its historical connotations, could be erroneously considered as an anomaly in the mod em United States.

Bubonic plague, the most common clinical form of the disease, is an acute, systemic febrile ifiness. The only specific feature, the bubo, is exquisitely tender, usually unilateral, and often has overlying

TABLE 3. SelectedFeaturesof ClinicalCoursein 31 Surviving Plague Patients New Mexico 1970 to 1980*

Feature of Clinical Course Time in Days@

erythema and perinodal edema. The bubo is most frequently inguinal, femoral, axifiary, or cervical, but can occur in any lymph node region. Buboes or bubo-associated pain and tenderness were reported as initial symptoms by 68% of bubonic patients; in the remainder, a typical bubo developed by the time of hospitalization. However, despite the bubo, plague entered the differential diagnosis at the first medical contact in only three patients and at the time of hospitalization in 13 patients.

Primary septicemic plague in children was less

common than the bubonic form of disease, but was

significantly more likely to result in both plague

pneumonia and death. An acute febrile ifiness with no pathognomonic features, primary septicemic plague was difficult to diagnose promptly. Given an average of only four days from onset to death for the five primary septicemic patients who died, along with an average delay of 3.7 days from onset of ifiness until administration of an effective antibiotic for all patients, the high case-fatality ratio of pri

mary septicemic plague seems difficult to lower.

Patients with primary septicemic plague may bene fit from a high index of clinical suspicion, close follow-up after an initial medical contact, and inclu

sion of an aminoglycoside antibiotic as part of broad spectrum therapy for community-acquired sepsis of unknown origin.

Plague pneumonia may occur secondary to hem atogenous spread or as a primary pneumonic infec tion from exposure to a person or animal with pneumonic plague.4 In this series, suspected or con

firmed plague pneumonia was associated with a

delay in initiation of effective antibiotic therapy and with primary septicemic plague. No cases of

primary pneumonic plague occurred. Although per

son-to-person spread of pneumonic plague has not been documented in this country since 1924,' the potential for rapid spread of highly virulent primary pneumonic plague mandates aggressive public health measures whenever plague pneumonia oc

curs.4'5

Cases of plague meningitis reported in this series are unusual, as only one occurred as a late compli

cation of bubonic or septicemic disease following

treatment with an ineffective antibiotic such as

penicillin.6'7

Recognizing plague or considering the diagnosis appears difficult, even in a state reporting two thirds of all plague cases in the United States.6 Unfortu nately, histories for contact with specific animals or fleas are uncommon for plague patients8 and the

physical finding of presumed insect bites was lim

ited to one third of patients. Although an increased risk of sylvatic plague acquisition has been epide

miologically linked with living in a “¿rodentattrac

* One surviving patient was not hospitalized and is not

included.

t Timeuntilmarked

symptomatic

improvement

anduntil

fever lysis is measured from start of administration of an

effective antibiotic; time until incision and drainage is

measured from onset of illness.

t Datea marked

increase

inwell-being

notedinhospital

chart; data available for 26 patients.

§

First 24-hourperiodduringwhichtemperature

re

mained below 37.8 C; data available for 27 patients.

ItIncisionanddrainageof buboperformedonnineof 29

tive― home and failing to dust pet dogs and cats with flea powder in rural areas of New Mexico,8 this type of information may not be available or sought by the physician. In addition, given the multiplicity of possible exposures to sylvatic plague in an en zootic rural environment (fleas acquired directly from wild rodents, from the environment, or carried on dogs and cats; direct contact with rodents, rab bits, or other plague-infected animals, including cats), the sensitivity and specificity of any host or vector contact data are uncertain. Therefore, resid ing in or visiting a rural area of the American West (especially from June through September) during the week prior to the onset of ifiness may often be the only useful epidemiologic clue to the diagnosis of plague.9

Rapid diagnostic procedures are necessary in the management of patients with suspected plague. Gram, Wayson's, or Giemsa stains of clinical speci mens, most commonly bubo aspirates, will reveal

small, bipolar-staining, Gram-negative coccobacilli. FA staining of additional slides can be completed within hours in a properly equipped laboratory.

Final confirmation of plague requires growth of the organism from clinical material or serologic diag nosis.4'5 Culture of Y pestis on blood agar often takes 48 hours or longer, as the organism is slow growing and grows optimally at 28 C. Serologic testing is only useful for retrospective diagnosis; the convalescent specimen should be obtained three to four weeks after onset.4

Although recovery from plague without antibiotic therapy is documented,'° the four patients in this series who were not given appropriate antibiotics died, and all patients who recovered received at least one effective antibiotic. Streptomycin is the

drug of choice for severe plague infection, including primary septicemic plague and plague pneumonia.4

It should be given in a dosage of 20 to 30 mg/kg/ day intramuscularly in two divided doses. Amino glycoside antibiotics other than streptomycin and kanamycin are probably effective, but extensive clinical experience with these drugs is lacking. Tet racycline (tetracycline hydrochloride, oxytetracy clime, chlortetracycline) may be given to patients with uncomplicated plague, preferably 8 years of age and older. Chioramphenicol is the drug of choice for meningeal plague and is a useful adjunct or alternative to other agents.

Trimethoprim-sulfamethoxazole has been used to treat plague,4 but has not been adequately eval uated to recommend its routine use. Penicfflins, semisynthetic penicillins, and cephalosporins are considered ineffective or partially effective against plague, despite variable in vitro susceptibility of the organism4 (J. Poland, personal communication,

1981). The total duration of antibiotic therapy should be at least two to four days after fever lysis.

Patients with plague recovered slowly, whether the imprecise measure of improved well-being or fever curves were used to assess clinical response following treatment. The average of six days from the start of effective antibiotic therapy until fever lysis and the need to incise and drain nine patients' buboes (31% of surviving bubonic patients) after more than two weeks of ifiness ifiustrate the slow resolution of plague signs and symptoms.

Three considerations highlight the current im portance of pediatric plague. First, the number of plague cases in this country is increasing; more cases were reported from 1970 to 1980 (121) than

during the preceding 40 years (1930 to 1969; 111

cases).―2 Since the first case of human plague was diagnosed in this country in 1899,' sylvatic plague involving wild rodents has been documented west of the 100th meridian, in counties from 15 western states.9 This large reservoir of enzootic plague in a rapidly growing section of the United States prob ably ensures a continuing increase in cases of hu man plague in this country.

The second factor contributing to plague as a pediatric concern is the significant increase in the proportion of plague cases occurring among chil then 16 years of age and younger. From 1950 to

1979, 55% of plague cases were in children, com

pared with only 24% during the period 1900 to 1949.' Third, the increase in plague and the proportion of pediatric cases is important to physicians and public health officials throughout the United States. The potential for travel during the incuba tion period or after the onset of illness, with an

accompanying threat both to the patient as a result

of delayed diagnosis and to public health if plague pneumonia supervenes, has been documented.9 Four of the 38 pediatric plague patients (11%) in this series traveled in this manner, and their disease was eventually diagnosed in California (fatal plague

pneumonia), Arizona, Montana, and Nebraska.

Local, state, and federal public health agencies are important sources of consultation and labora tory assistance for pediatricians with suspected plague patients. Prompt reporting of these patients to health departments is also vital to protect the public health from potential consequences of this medieval-sounding disease whose natural reservoir is now firmly embedded in the American West.

ACKNOWLEDGMENTS

The authors thank George Schmid, MD, for review of

the manuscript, Ted Brown for environmental data, and

the physicians of New Mexico for their continued coop

REFERENCES

1. Link VB: A History of Plague in the United States of

America. US Public Health Service monograph no. 26,

Washington, DC, 1955, pp 1-120

2. Kaufmann AF, Boyce JM, Martone WJ: Trends in human plague in the United States. J Infect Dis 141:522, 1980 3. Alsofrom DJ, Mettler FA, Mann JM: Radiographic manifes

tations of plague in New Mexico, 1975-80. Radiology 139:561, 1981

4. Poland JD, Barnes AM: Plague, in Steele JH (ed): CRC

Handbook Series in Zoonoses. Boca Raton, FL, CRC Press

Inc, 1979, vol 1, pp 515—559

5. Bahmanyar M, Cavanaugh DC: Plague Manual. Geneva,

World Health Organization, 1976, pp 1-76

6. Reed WP, Palmer DL, Williams RC, et al: Bubonic plague in the southwestern United States: A review of recent ex perience. Medicine 49:465, 1970

7. MartIn AR, Hurtado FP, Plessals HA, et al: Plague menin gitis: A report of three cases in children and review of the

problem. Pediatrics 40:610, 1967

8. Mann JM, Martone WJ, Boyce JM, et al: Endemic human

plague in New Mexico: Risk factors associated with infection.

J Infect Dis 140:397, 1979

9. Mann JM, Schmid GP, Stoesz PA, et al: Peripatetic plague. JAMA 247:47, 1982

10. Pollitzer R: Plague. Monograph Series no. 225. Geneva,

World Health Organization, 1955, pp 1-698

FRIEDREICH'S ATAXIA

The clinical features of 115 patients from 90 families with Friedreich's ataxia are described. Onset of symptoms was before the age of 25 (mean 10.52) years in all the index cases. An analysis of early cases suggested that limb and truncal ataxia and absent tendon reflexes in the legs were the only consistent diagnostic criteria within five years of presentation. Dysarthria, signs of pyramidal tract dysfunction in the legs and loss of joint position and vibration sense are not necessarily present during the first five years of symptoms, but appear to develop eventually in all cases. Scoliosis and ECG evidence of cardiomyopathy were found in over two-thirds of the patients studied; pes cavus, distal amyotro phy, optic atrophy, nystagmus and deafness were all less frequent. The disorder was gradually progressive in all cases. The mean age oflosing the ability to walk was 25 years; 95 per cent were chair-bound by the age of 44 years.

About 10 per cent of the patients had diabetes mellitus which was controlled by oral hypoglycaemic drugs in one quarter. Diabetes appeared to be associated with a higher incidence of optic atrophy and deafness. Diabetes also clustered within sibships; the risk of an individual with Friedreich's ataxia developing diabetes if an affected sib has it is over 40 per cent. Similarly, cardiomyopathy ran true within affected members of the same sibship, but there were instances of discordance which suggest that the development of the non-neurological features of Friedreich's ataxia may be controlled by modifying genes rather than heterogeneity of the main gene.

Comment: This is the largest series ever reported and there are excellent data on early features, prognosis, and associated medical problems such as diabetes and cardiomyopathy. This information would be of value to the pediatrician in counseling parents of affected children.

Stephen C. Copps, MD

1982;69;762

Pediatrics

Jonathan M. Mann, Laurence Shandler and Alice H. Cushing

Pediatric Plague

Services

Updated Information &

http://pediatrics.aappublications.org/content/69/6/762 including high resolution figures, can be found at:

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtml entirety can be found online at:

Information about reproducing this article in parts (figures, tables) or in its

Reprints

1982;69;762

Pediatrics

Jonathan M. Mann, Laurence Shandler and Alice H. Cushing

Pediatric Plague

http://pediatrics.aappublications.org/content/69/6/762

the World Wide Web at:

The online version of this article, along with updated information and services, is located on

American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.