Childhood Fever: Correlation of Diagnosis With Temperature Response to Acetaminophen

(1)

Childhood

Fever:

Correlation

of Diagnosis

With

Temperature

Response

to Acetaminophen

M. Douglas

Baker,

MD, Patricia

D. Fosarelli,

MD, and

Richard

0. Carpenter,

MD

From the Department of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, Maryland

ABSTRACT. Many people believe that temperature

re-sponse to antipyretics in febrile children varies according

to diagnosis. To evaluate the validity of this premise, we

prospectively studied the temperature response to

acet-aminophen of febrile children who came to an urban

pediatric emergency and walk-in facility. The study group

consisted of 1,559 patients between the ages of 8 weeks

and 6 years whose temperatures when seen were greater

than 38.4’C and who had not received antipyretic treat-ment within the previous four hours. Acetaminophen (15 mg/kg) was administered to each child and repeat tern-peratures were taken one and two hours later. Patient management was unaffected by the study, and physicians

were unaware of the repeat temperature measurements.

Telephone follow-up was conducted with the parents of

each child within five days of the initial visit. Children with cultures positive for bacterial disease or chest x-ray

films positive for pneumonia had slightly greater one-and two-hour temperature decreases compared with chil-dren with other diagnoses. Although statistically signifi-cant, we do not consider these differences in response to be clinically useful. We conclude that fever response to

acetaminophen is not a clinically useful indicator by

which to differentiate the causes of febrile illnesses in

young children. Pediatrics 1987;80:315-318; fever, acet-aminophen, antipyretic.

In the mid-l9th century, Claude Bernard com-pleted his experiments on the overheating of

ani-mals, proving that death quickly occurred if the body’s temperature increased 5#{176}Cto 6#{176}Cabove normal.1 Thereafter, fever was generally regarded as injurious to health, and treatment with

antipy-Received for publication July 14, 1986; accepted Nov 7, 1986. Presented, in part, at the annual meeting of the Ambulatory Pediatric Association, Washington, DC, May, 1986.

Reprint requests to (M.D.B.) Emergency Department, The Chil-dren’s Hospital of Philadelphia, 34th St and Civic Center Blvd, Philadelphia, PA 19104.

retic medications was considered imperative. More modern studies have demonstrated some beneficial

effects of increased body temperatures.28 However, the role of fever as an essential defense mechanism remains unclear.

Although modern physicians seem to have achieved a more balanced perspective on the

treat-ment of fever, many parents appear to be as fearful of fever as physicians were a century ago. In one

recent study parental misconceptions about the

dangers of fever were investigated.9 Most parents

interviewed were unduly worried about low-grade fever, ie, temperatures less than 38.9’C, and slightly more than one half (52%) believed that tempera-tures less than 40#{176}Ccould cause serious neurologic

side effects. Hence, most parents aggressively

treated fever; 85% gave antipyretics for

tempera-tures of 38.4#{176}Cor less.

One seemingly common concern shared by

par-ents and physicians alike is the significance of fevers that are not relieved by antipyretic

treat-ment. Although it seems appropriate that less-re-sponsive fevers should be predictive of more serious

disease, no reports have corroborated this

hypoth-esis. In fact, one investigation of children younger

than 2 years of age with temperatures of 38.9#{176}Cor

greater who were being evaluated for occult

bacte-remia showed no difference between children with bacteremia and children without bacteremia in av-erage decrease in temperature after antipyretic

administration.’0 The current study was designed

to prospectively investigate whether or not

temper-ature response to acetaminophen administration varies by disease process.

METHODS

(2)

316 FEVER RESPONSE

clinic for evaluation of febrile illnesses. Only

chil-dren whose rectal temperatures were greater than

38.4#{176}Cand who had received no antipyretics within

the previous four hours were eligible for entry into the study. After initial temperature measurement, each child received oral acetaminophen, 15 mg/kg.11 Children vomiting within 30 minutes of receiving acetaminophen were remedicated with a second

full dose. Patient evaluation and management by

pediatric residents then proceeded unaffected by the study. Repeat temperature measurements were made one and two hours after acetaminophen administration (exact times were recorded by the

nursing staff). However, patients whose evaluations

were completed prior to one or two hours were not detained for purposes of measuring temperatures. Temperature measurements were obtained using electronic thermometers with

digital

read-out.

Demographic information and vital signs were recorded by the nursing staff. Residents were not made aware of the one- and two-hour temperature

measurements. Patient charts were subsequently

reviewed by the primary investigator (M.D.B.), and information concerning the diagnosis, pertinent historical and physical findings, laboratory data, and disposition were recorded. Telephone contact was made with each patient’s parents within five

days of the initial visit, both to check the patient’s

clinical status and to ascertain whether the child had received interim medical care altering the

orig-inal diagnosis. The study was conducted for 1 year,

from October 1984 to September 1985. Statistical

analysis was performed using x2 and analysis of variance.’2 Significance was defined as

P

< .05. The

study protocol was approved by the Joint

Commit-tee on Clinical Investigation of the Johns Hopkins University School of Medicine and the Johns

Hop-kins Hospital. Informed consent was not obtained.

RESULTS

A total of 3,911 patients within the specified age

range were seen during the study period for evalu-ation of febrile illnesses. Of these children, 47% were premedicated and thus ineligible for enroll-ment. Of the 2,055 eligible patients, 76 (4%) were

erroneously omitted and 420 (20%) were evaluated

and discharged within one hour of the

acetamino-phen treatment. These 496 patients were thus

ex-cluded from the study. Of the remaining 1,559

patients, all had repeat temperature measurements

one hour after acetaminophen treatment and 471

(30%) had two-hour temperatures recorded. As

in-dicated by the times recorded, all repeat tempera-ture measurements were made within five minutes of the proposed one- and two-hour target intervals.

Boys outnumbered girls 850 to 709. Racial distri-bution was 1,344 black, 213 white, and two oriental.

The age distribution of the patients was as follows:

less than 12 months of age, 34%; 12 to 23 months

of age, 22%; 24 to 35 months of age, 17%; 36 to 47 months of age, 11%; 48 to 59 months of age, 8%; 60

to 71 months of age, 5%; and 72 to 83 months of age, 3%.

A total of 46 different diagnoses were assigned to

study participants. These were grouped into seven

diagnostic categories. The miscellaneous group, which included children with “viral syndrome” or upper respiratory tract infection, was 30% of the total. Otitis media (27%), viral diseases (15%),

chest x-ray film-positive pneumonia (11%),

non-cultured gastroenteritis ( 10%

),

culture-positive

bacterial disease (4%), and group A j3-hemolytic

Streptococcus pharyngitis (3%) accounted for the

remainder. All patients in the lattermost group (group A f-hemolytic Streptococcus pharyngitis)

were older than 22 months of age. The viral disease group included patients with coxsackie hand-food-and-mouth disease, varicella, croup,

gingivostoma-titis, mononucleosis, and those with positive viral

enzyme-linked immunosorbent assays or cultures negative for bacteria. Culture-positive bacterial

dis-eases included urinary tract infection, sepsis, men-ingitis, cellulitis, osteomyelitis, septic arthritis,

si-nusitis, adenitis, pyelonephritis, peritonitis, and

felon.

One-hour temperature responses are listed in Table 1. Children with group A 3-hemolytic

Strep-tococcus pharyngitis, other culture-positive

bacte-rial diseases, or chest x-ray-film-positive pneu-monia had larger temperature decreases than those

in the remaining diagnostic groups. Although small, these differences were statistically significant

(analysis of variance:

P

< .01). Similar trends were seen in the two-hour temperature responses (Table

1). Children with culture-positive bacterial diseases

or chest x-ray film-positive pneumonia again

dem-onstrated greater temperature decreases than chil-dren with other diagnoses

(P

< .01). There were no differences in either one- or two-hour temperature changes associated with age, sex, or race. Within each diagnostic group, the initial temperatures of the patients with two-hour temperature

measure-ments

did

not differ significantly from those having

only one-hour temperature measurements.

Temperature data from children with bacterial deep tissue infections are presented in Table 2. Ten

patients had sepsis (eight Streptococcus pneumo-niae, one group B Streptococcus, and one Esche-richia coli), five had meningitis (Haemophilus

influ-enzae), five had ventriculoperitoneal shunt infec-tion (four with Staphylococcus epidermidis and one

by guest on September 7, 2020 www.aappublications.org/news

(3)

TABLE 1. One- and Two-Hour Tern perature Responses*

Diagnostic Category Initial Temperature (n = 1,559)

1-h Change (n = 1,559)

2-h Change (n = 471) Group A f-hernolytic Streptococcus 39.3 ± 0.5 1.3 ± 0.51 1.4 ± 0.4

pharyngitis

Bacterial diseases 39.7 ± 0.8 1.3 ± 0.8t 1.8 ± 0.5t

Gastroenteritis 39.5 ± 0.6 1.1 ± 0.6 1.4 ± 0.7

Pneumonia 39.6 ± 0.7 1.2 ± 0.6t 1.8 ± 0.6t

Viral diseases 39.6 ± 0.6 1.0 ± 0.6 1.4 ± 0.7

Otitis media 39.6 ± 0.6 1.0 ± 0.6 1.5 ± 0.7

Miscellaneous 39.5 ± 0.4 1.0 ± 0.6 1.6 ± 0.7 Total (N = 1,559) 39.5 ± 0.6 1.0 ± 0.6 1.6 ± 0.7 *Ternperatures are ‘C ± SD.

t P < .01, analysis of variance.

TABLE 2. Temperature Responses in Children With Bacte rial Deep Tissue Infections Diagnosis (No.) Initial

Temperature

(Mean ‘C)

1-h 2-h

Change Change

(Mean ‘C) (Mean ‘C)

Sepsis (10) 40.1

Meningitis (5) 39.5

Shunt infection (5) 39.7 Septic arthritis (4) 39.1

Osteornyelitis (2) 39.4

Peritonitis (1) 40.1

Pyelonephritis (1) 38.8 Retropharyngeal abscess (1) 40.1

1.5 1.8

1.1 1.1

1.3 1.8

1.3 1.6

1.3 2.6

1.0 1.6

1.6 2.9

0.1 1.3

TABLE 3. Least and Greatest Temperature Responses

Diagnosis (No.) No. (%) of Patients With 1-h Decrease in Temperature <0.5’C

No. (%) of Patients With 1-h Decrease in Temperature >1.5’C

Group A f-hemolytic Streptococcus 4 (8) pharyngitis (40)

Bacterial diseases (61) 3 (5)*

Gastroenteritis (160) 28 (18)

Pneumonia (178) 13 (7)*

Viral diseases (225) 28 (12)

Otitis media (413) 73 (18)

Miscellaneous (473) 95 (20)*

Total (1,559) 244 (16)

17 (35)*

16 (26) 38 (24) 59 (33)*

24 (11)* 80 (19) 94 (20) 328 (21)

*P< .001, x2.

with S pneumoniae), four had osteoarthritis (two with Spneumoniae, one with Staphylococcus aureus and one with

H influenzae),

two had osteomyelitis (S aureus), and there was one each with peritonitis (multiple bowel flora), pyelonephritis

(E coli),

and

retropharyngeal abscess (group A Streptococcus). Statistical analysis was not performed on these groups because of the small number of patients in each diagnostic category. However, of those di-agnostic groups having more than one patient, meningitis seemed to show the least temperature response to acetaminophen administration. All

pa-tients with bacterial deep tissue infections

demon-strated a temperature decrease of at least 1#{176}C within two hours after receiving acetaminophen.

Data for patients with the greatest and least temperature changes one hour following

acetami-nophen administration are given in Table 3. A lower

percentage of those with pneumonia and other cul-ture-positive bacterial diseases responded to acet-aminophen with a decline in temperature of less than 0.5#{176}C.The diagnostic group with the highest percentage of patients with the least temperature response to acetaminophen was the miscellaneous

group. On the other hand, a significantly greater

proportion of children with group A 13-hemolytic

Streptococcus

pharyngitis and chest x-ray fllm-pos-itive pneumonia had a decrease greater than 1.5#{176}C. The lowest percentage of patients showing this response was in the viral group.

DISCUSSION

(4)

318 FEVER RESPONSE

association that might exist between diagnosis and

fever responsiveness to acetaminophen. It was our expectation that temperature changes would vary

little between disease processes, and our data have

shown this to be true. Although statistically signif-icant differences in both one-hour and two-hour temperature changes have been demonstrated,

these differences are too small to be of clinical usefulness. The bacterial diseases showed the great-est overall one- and two-hour responses,

undermin-ing the premise that serious illnesses are less

re-sponsive to antipyretics than more benign illnesses.

Even bacterial meningitis (H influenzae), although

less responsive to antipyretics than other bacterial diseases, showed an average temperature decrease

of more than 1#{176}Cone and two hours after acetami-nophen administration.

We recognize that our results may have been

biased by the clinical nature of many of the diag-noses assigned. Residents’ use of laboratory tests

to establish these diagnoses varied considerably.

Although attending physicians precepted resident management of patients, there assuredly existed some variability of the rigor with which various diagnoses were sought. We attempted to minimize

the number of misdiagnoses by making telephone contact with patients’ parents several days after each visit and confirming any alterations in

diag-noses that occurred during that time. However, it

is conceivable that some study children may have

been misdiagnosed and thus misassigned to their

diagnostic groups. Moreover, although our study

group was large, the exclusion of the 496 potential

study participants who were discharged within one hour of registration might also have affected our results. These children could possibly have been

those who were least ill-appearing and perhaps

most responsive to antipyretics. Nevertheless, our data show that patients with a variety of known

viral and bacterial diseases demonstrate similar

ranges of temperature response to acetaminophen

treatment.

Our data show that temperature responsiveness

to antipyretic administration cannot be reliably

used to distinguish between different diagnoses. This point should be kept in mind when evaluating

febrile patients in any setting, particularly when

doing telephone triage. Our undocumented

experi-ence is that physicians and parents seem to worry less about fevers that significantly decrease follow-ing acetaminophen treatment than they do about unresponsive or minimally responsive tempera-tures. Yet, we found that children with serious

bacterial diseases requiring expeditious treatment

often show the greatest temperature response to antipyretics.

Clinical investigators have long tried to isolate

clinical and laboratory data that help to identify specific diseases. Regrettably, these factors, either alone or in combination, often lack the sensitivity and specificity required for clinical usefulness. Our experience indicates that response to antipyretic

therapy does not help to distinguish between dis-ease states in young febrile children. At present,

clinical assessment by an experienced examiner, accompanied by the prudent use of laboratory tests, remains the best guide to the management of febrile children.

REFERENCES

1. Akerren Y: On hyperpyretic conditions during infancy and childhood. Acta Paediatr 1943;31:1-72

2. Bennett I, Nicastri A: Fever as a mechanism of resistance.

Bacteriol Rev 1960;24:16-34

3. Katersky J, Kass E: Is suppression of fever or hypothermia useful in experimental or clinical infectious disease? J Infect Dis 1970;121:81-86

4. Phelps A, Stanislaw D: Polymorphonuclear leukocyte mo-bility in vitro: Effect of pH, temperature, ethylalcohol, and caffeine, using a modified Boyden Chamber technique. Ar-thritis Rheum 1969;12:181-188

5. Sebag J, Reed WP, Williams RC Jr: Effect of temperature on bacterial killing by serum and by polymorphonuclear leukocytes. Infect Immun 1977;16:947-954

6. RObertS NJ, Steigbigel RT: Hyperthermia and human

leu-kocyte functions: Effects on response of lymphocytes to mitogen and antigen and bactericidal capacity of monocytes and neutrophils. Infect Immun 1977;18:673

7. Smith JF, Knowlton RP, Agarwal 55: Human lymphocyte responses are enhanced by culture at 40 degrees C. J. Im-munol 1978;121:691-694

8. Heron I, Berg K: The actions of interferon are potentiated by elevated temperature. Nature 1978;174:508-510

9. Schmidt B: Fever phobia Am J Dis Child 1980;134:176-181 10. Torrey SB, Henretig F, Fleisher G, et al: Temperature

response to antipyretic therapy in children: Relationship to occult bacteremia. Am J Emerg Med 1985;3:190-192

11. Done AK: Treatment of fever in 1982: A review. Am J Med,

June l983,pp 27-34

12. Zai J: BiostatisticalAnalysis. Englewood Cliffs, NJ, Prentice Hail Inc, 1974, pp 60-68

(5)

1987;80;315

Pediatrics

M. Douglas Baker, Patricia D. Fosarelli and Richard O. Carpenter

Acetaminophen