from Paul L. McCarthy, Kliegman: Nelson Textbook of Pediatrics, 18th ed.
There are many reasons for a sick child visit, but most are due to acute self-limited intercurrent infections; often the child is febrile. When evaluating an acutely ill, febrile child, the pediatrician must be aware of categorical statistics about the probable occurrence of serious illness, because one of the major goals of the sick child visit is to identify the seriously ill child who requires specific therapeutic intervention. The risk for and the cause of serious illness in the acutely febrile child vary, depending on age. The infant in the 1st 3 mo of life is more susceptible to sepsis and meningitis caused by group B streptococci and gram-negative organisms. Infants in the 1st mo of life are at highest risk. Urinary tract infections are more frequent in males; infants in this age group more often have an underlying anatomic abnormality of the urinary tract than do older children with urinary tract infections. As the infant matures beyond 3 mo, the bacterial pathogens that usually cause bacteremia, sepsis, and meningitis are Streptococcus pneumoniae, Haemophilus influenzae type b (if the child is unimmunized or only partially immunized), and Neisseria meningitidis. Immunization against some serotypes of S. pneumoniae may reduce the occurrence of occult bacteremia and serious infections caused by that organism, as has immunization against H. influenzae type b. After infancy, urinary tract infections are seen more often in females. Immunity develops rapidly to the common bacterial pathogens during the first 3–4 yr of life. N. meningitidis is the leading cause of bacterial meningitis. In children older than 36 mo, pharyngitis caused by group A streptococci is a common bacterial infection. Mycoplasma pneumoniae assumes increasing importance as a cause of pulmonary infiltrates in children older than 5 yr of age. Table 1 shows serious illnesses documented in children in the 1st 3 yr of life who presented with fever and acute illness at a university hospital and private practice. In many studies, urinary tract infections are the most common serious bacterial infections. Soft tissue infections due to streptococcus or staphylococcus may include cellulitis, fasciitis, osteomyelitis, and septic arthritis. Noninfectious, but serious disease should also be considered and include trauma (abuse), midgut volvulus, appendicitis, intussusception, poisoning (salicylates), metabolic disorders (hypoglycemia, hyperammonemia), neurologic disorders (seizures, infant botulism), or inflammatory diseases (
The acutely ill child with a serious illness is identified by careful observation, history taking, physical examination, appreciation of age and body temperature as risk factors, and the judicious use of screening laboratory tests. The physician can use the data to make informed decisions about the need for more definitive laboratory tests (urine culture), therapy, and the advisability of hospital admission. Observation, history, and physical examination are integrated when the sick child evaluation is being done; that is, as the child is being observed, historical data are gathered. History taking and observational assessment often continue as the physical examination is performed. If abdominal tenderness is found on examination, additional history about blood in the stool, cramping abdominal pain, and vomiting may be sought.
TABLE 1 -- Diagnosis of Serious Illnesses During 996 Episodes of Acute Infectious Illness in Febrile Children Younger Than 36 Mo[*]
| | CASES | |
| DIAGNOSIS | NO. | % |
| Bacterial meningitis | 9 | 0.9 |
| Aseptic meningitis | 12 | 1.2 |
| Pneumonia | 30 | 3.0 |
| Bacteremia | 10 | 1.0 |
| Focal soft tissue infection[†] | 10 | 1.0 |
| Urinary tract infection | 8 | 0.8 |
| Bacterial diarrhea | 1 | 0.1 |
| Abnormal electrolytes or blood gases | 9 | 0.9 |
| Total | 89 | 8.9 |
From
| * | Includes cellulitis, osteomyelitis, and septic arthritis. |
| † | Includes cellulitis, osteomyelitis, and septic arthritis. |
Observation is important in the evaluation of the acutely ill child. The child should be observed for specific evidence of a serious illness, such as grunting, which might indicate pneumonia or sepsis, or a bulging fontanel, which might indicate bacterial meningitis or head trauma. Most observational data that the pediatrician gathers during an acute illness should focus, however, on assessing the child's response to stimuli. How does the crying child respond to the parents' comforting? How quickly does the sleeping child awaken with a stimulus? Does the child smile when the examiner interacts with him or her? Assessing responses to stimuli, while often providing those stimuli, requires knowledge of normal responses for different age groups, the manner in which those normal responses are elicited, and to what degree a response might be impaired.
Sometimes the manner in which the child responds to stimuli is readily apparent. For example, the child may vocalize and smile as the examiner enters the room. At other times, more effort and more stimuli are needed to cause the child to act in a normal manner. Often the fussing, irritable child begins to look around and focus on the examiner when held and walked by the parent. This normal visual behavior is an important indicator of well-being. Thus, during observation, the pediatrician must be both clinically and developmentally oriented.
Six observation items and their scales (Acute Illness Observation Scales) that have reliably and validly identified serious illness in febrile children are shown in Figure 1 . A normal finding is scored as 1, moderate impairment as 3, and severe impairment as 5. The best possible score is 6 items × 1 = 6; the worst score is 6 items × 5 = 30. The chance of serious illness is 1–2% if the total score is 10 or less; if the score is >10, the risk of serious illness increases by at least 10-fold. It is not clear whether these scales can be used in the first 1–3 mo of life because infants may not have developed the skills required to score some of these items.
Figure 1 Acute illness observational scales for use in clinical evaluation of the well and sick child. (From
HISTORY
History taking is complex. Parents must transmit how a younger child has been “feeling.” Parents should also provide information on a specific symptom, such as bloody diarrhea or cyanosis when coughing. The older child's perception of symptoms may reflect a developmentally immature understanding of causation. The examiner pursues the historical information provided by the parents or child to define the symptoms precisely. If the complaint is blood in the stool, additional questions can be asked about other evidence of bowel inflammation, such as watery stools, mucus in the stools, or increased frequency of stooling. If the historical information indicates crying with defecation and streaks of blood on the outer portion of a hard stool, without other changes in the character or frequency of the stool, a diagnosis of a rectal fissure is tenable.
Questions should focus on those entities that are seen most commonly in acute febrile childhood illnesses. The more serious diagnoses are outlined in Table 1 . Organ-specific questions are helpful, such as those concerning fast breathing, cyanosis, retraction, and wheezing with pneumonia, or pain, swelling, and pseudoparalysis with septic arthritis. Because most acute illnesses in children are caused by minor viral infections, specific questions about the epidemiology of the illness can offer important insights. Are there other children in the family with similar symptoms? Has the child had other illness exposures? Finally, it is important to be aware of any underlying chronic problems that might predispose the child to recurring infections or a serious acute illness; for example, the child with sickle cell anemia or AIDS is at increased risk for recurrent episodes of bacteremia. Benign viral illnesses may produce serious secondary consequences, such as dehydration or severe respiratory syncytial virus pneumonia in infants who had bronchopulmonary dysplasia as neonates. Additional questions to assess hydration, such as questions about wet diapers, tears, or awakeness, should be used to assess secondary complications.
PHYSICAL EXAM
During physical examination, the pediatrician seeks evidence of illnesses, especially serious illnesses (see Table 1 ), that are causes of acute febrile episodes in children. The portions of the physical examination that require the child to be optimally cooperative are completed first. Initially, it is best to seat the child on the parent's lap; the older child may be seated on the examination table. Vital signs are often overlooked, but are valuable in assessing ill children. The degree of fever, the presence of tachycardia out of proportion to the fever, and the presence of tachypnea and hypotension all suggest a serious infection. In addition to the general level of interaction, color, and hydration, as assessed by the Acute Illness Observation Scales (see Fig. 1 ), the child's respiratory status is evaluated. This evaluation includes determining respiratory rate and noting any evidence of inspiratory stridor, expiratory wheezing, grunting, or coughing. Evidence of increased work of breathing—retraction, nasal flaring, and the use of abdominal musculature—is sought. Because acute infections in children are most often caused by viral infections, the presence of nasal discharge is noted. It is possible at this time to assess the skin for rashes. Frequently, viral infections cause an exanthematous eruption, and many of these eruptions are diagnostic (the reticulated rash and “slapped-cheek” appearance caused by parvovirus infections or the typical appearance of hand-foot-and-mouth disease caused by coxsackieviruses). The skin examination may also yield evidence of more serious infections (bacterial cellulitis or petechiae associated with bacteremia). Cutaneous perfusion should be assessed by warmth and capillary refill time. When the child is seated and is least perturbed, an assessment of fontanel tension can be completed; it can be determined if the fontanel is depressed, flat, or bulging. It is also important to assess the child's willingness to move and ease of movement. Usually, the child with meningitis will hold the neck stiffly and will often cry when any attempt is made to flex the neck, even during cuddling by the parent. This is termed paradoxic irritability. The child with cellulitis, osteomyelitis, or septic arthritis in an extremity will resist movement of that limb. The child with peritoneal inflammation will sit quietly and become irritable during movement. It is reassuring to see the child moving about on the parent's lap with ease and without discomfort.
During this initial portion of the physical examination, when the child is most comfortable, the heart and lungs are auscultated. In the acutely febrile child, because of the relatively frequent occurrence of respiratory illnesses, it is important to assess adequacy of air entry into the lungs, equality of breath sounds, and evidence of adventitial breath sounds, especially wheezes, rales, and rhonchi. The coarse sound of air moving through a congested nasal passage is frequently transmitted to the lungs. The examiner can become attuned to these coarse sounds by placing the stethoscope near the child's nose and then compensating for this sound as the chest is auscultated. The cardiac examination is completed next; findings such as pericardial friction rub, loud murmurs, or distant heart sounds may indicate an infectious process involving the heart. The eyes are examined to identify features that might indicate an infectious process. Often, viral infections result in a watery discharge or redness of the bulbar conjunctivae. Bacterial infection, if superficial, results in purulent drainage; if the infection is more deep-seated, tenderness, swelling, and redness of the tissues surrounding the eye are present, as well as proptosis, reduced visual acuity, and altered extraocular movement. The extremities may then be evaluated not only for ease of movement, but also for the possibility of swelling, heat, or tenderness; such abnormalities may indicate focal infections.
The components of the physical examination that are more bothersome to the child are completed last. This is best done with the patient on the examination table. Initially, the neck is examined to assess for areas of swelling, redness, or tenderness, as may be seen in cervical adenitis. The neck is then flexed to evaluate suppleness; resistance to flexion is indicative of meningeal irritation. The Kernig and Brudzinski signs may be sought at this time. In children younger than 18 mo, meningeal signs may not always be present with meningitis; however, if they are present, the diagnostic implications are the same as for the child older than 18 mo. During examination of the abdomen, the diaper is removed. The abdomen is inspected for distention. Auscultation is performed to assess adequacy of bowel sounds, followed by palpation. The child often fusses as the abdomen is auscultated and palpated. Every attempt should be made to quiet the child; if this is not possible, increased fussing as the abdomen is palpated may indicate tenderness, especially if this finding is reproducible. In addition to focal tenderness, palpation may elicit involuntary guarding or rebound tenderness; these findings indicate peritoneal irritation, as is seen in appendicitis. The inguinal area and genitals are then sequentially examined. In the febrile child, inguinal adenitis or a strangulated hernia may be the cause of fever. The child is then placed in the prone position, and abnormalities of the back are sought. The spine and costovertebral angle areas are percussed to elicit any tenderness; such findings may be indicative of osteomyelitis or diskitis and pyelonephritis, respectively.
Examining the ears and throat completes the physical examination. These are usually the most bothersome parts of the examination for the child, and parents frequently can be helpful in minimizing head movement. During the oropharyngeal examination, it is important to document the presence of enanthemas; these may be seen in many infectious processes, such as hand-foot-and-mouth disease caused by coxsackievirus. This portion of the examination is also important in documenting inflammation or exudates on the tonsils, which may be viral or bacterial.
At times, repeating portions of the assessment is indicated. If the child cried continuously during the initial clinical evaluation, the examiner may not be certain if the crying was caused by the high fever, stranger anxiety, or pain, or is indicative of a serious illness. Continual crying also makes portions of the physical examination, such as auscultation of the chest, more difficult. Before a repeat assessment is performed, efforts to make the child as comfortable as possible are indicated. Such efforts include reducing the fever with antipyretics and offering the child a bottle. Because most children with fever do not have serious illnesses, repeated assessments are more likely to document normal findings. If, on the other hand, the child is persistently irritable, the possibility of serious illness increases.
RISK FACTORS
The sensitivity of the carefully performed clinical assessment, observation, history, and physical examination for the presence of serious illness is approximately 90%. Careful data gathering is necessary in the observation, history, and physical examination, because each component of the evaluation is as effective as the others in identifying serious illness. Other data, however, should be sought to improve this sensitivity level. In the child with an acute febrile illness, important supplemental data are age, body temperature, and the results of screening laboratory tests. Febrile children in the first 3 mo of life have yet to achieve immunologic maturity and therefore are more susceptible to severe infections and to infections by unusual organisms. Thus, the febrile infant is at greater risk for serious bacterial infection than the child beyond 3 mo of age. In febrile children, the higher the fever, the greater the risk of serious illness. The risk of bacteremia in infants increases as the degree of fever increases. The limit of physiologic thermoregulation is 41.1°C (106°F); fevers in this range and higher indicate bacteremia, but also possible central nervous system infection, pneumonia, or pathologic hyperthermia.
Screening laboratory tests may be helpful in identifying the febrile child at increased risk for selected serious illnesses. S. pneumoniae is currently the most common cause of occult bacteremia not associated with a focal soft tissue infection. A total white blood cell count of ≥15,000/mm3 and/or an absolute neutrophil count of ≥10,000/mm3, in addition to age 3–36 mo, higher grades of fever, and a more ill appearance, are indicators of increased risk for occult bacteremia caused by S. pneumoniae. The incidence of occult pneumococcal bacteremia in febrile children may be declining because of the introduction of conjugated pneumococcal vaccine. Urinalysis and urine culture must be considered when the source of fever is not apparent, especially in the highest-risk groups: females and uncircumcised males younger than 2 yr of age and all boys younger than 1 yr of age. The presence of leukocyte esterase, >5 white blood cells/high-power field on a spun urine specimen, or bacteria by Gram stain on an unspun urine specimen suggests urinary tract infection, but the sensitivity of these indicators is, on average, only 75–85% and urine culture is the definitive test. An elevated C-reactive protein value may also distinguish bacterial from viral infection.
MANAGEMENT
If the febrile child is older than 3 mo and appears well, if the history or physical examination does not suggest a serious illness, and if no age or temperature risk factors are present, the child may be followed expectantly. If otitis media is present, it should be treated. This profile applies to most children with acute infectious illnesses. If, on the other hand, the child appears ill or the history or physical examination suggests a serious illness, definitive laboratory tests appropriate for those findings are indicated (e.g., chest radiograph for a child with grunting). The area of greatest controversy is whether laboratory studies are needed in a febrile child who appears well and has no abnormalities on history and physical examination, but who is younger than 3 mo of age or whose temperature is high. Many would agree that a sepsis work-up is indicated in the febrile child younger than 1 mo and possibly younger than age 3 mo. Obtaining blood cultures in children older than 3 mo with higher grades of fever has gained increased acceptance.
If the physician feels comfortable in following as an outpatient the child in whom no specific diagnosis has been established, a follow-up examination often yields a diagnosis. During the initial visit, or from one visit to the next during the acute illness, the change in symptoms or in the findings on physical examination over time may provide important diagnostic clues. For the child in whom a diagnosis has already been established and who does not require hospitalization, follow-up by telephone or an office visit should be used to monitor the course of the illness and to further educate and support the parents.