Routine Measurement of Head Circumference as a Tool for Detecting Intracranial Expansion in Infants: What Is the Gain? A Nationwide Survey

METHODS

Norway is a sparsely populated country with a public, 3-level, hierarchical hospital structure, with local community hospitals as the primary referral centers. Most counties have a central hospital with a pediatric department as a secondary referral center. At the top, there are 5 university clinics with a neurosurgical department, each serving a health region consisting of several counties. Few counties have >1 pediatric department; the total number of such departments in Norway is 21. During the study period, the country's population increased from 4.3 million to 4.5 million. During the study period, each year a mean of 58 903 children (range: 55 434–60 927 children) were born and a mean of 300 416 children were <5 years of age.⁴ Norwegian children with intracranial expansions are all treated within the country.

This national study included all Norwegian children <5 years of age who were diagnosed as having an intracranial expansive condition between January 1, 1999, and December 31, 2002. The hospital records of all 21 Norwegian departments of pediatrics and the 5 departments of neurosurgery were searched for one of the following International Classification of Diseases, 10th Revision, diagnoses: C70 to C72, D33, D43, G91, G93 to G94, I60 to I64, or Q03 to Q05 (covering all intracranial tumors, cysts, congenital and acquired hydrocephalus, and hemorrhage). Only primary conditions were the objects of our investigations. Therefore, conditions in which secondary hydrocephalus is to be expected (eg, after closure of a myelomeningocele or after intracranial surgery) were not included. According to the instructions for health care personnel working in Norwegian “health stations” for children, all children with rapidly increasing HC should be referred to a specialist. Rapidly increasing HC is defined as crossing 2 percentile curves on the HC registration sheet, which is based on Norwegian reference values.^5–7 In almost all cases, such children are referred to the regional pediatric department and thus have medical records there. According to the same instructions, infants who are born with a large HC that follows the percentile curve are not to be referred unless they display clinical symptoms.

Instructions given by the Norwegian health authorities regulate the activities of the health centers. These instructions have the legal authority of law within the Norwegian medicolegal system. Therefore, it is mandatory for the parents to bring the child to the local health center at certain intervals. Norwegian recommendations are that HC is to be measured routinely at each regular visit to the health center during the first year of life; later, HC is measured when indicated.⁸

We visited all of the pediatric departments to collect information from the medical records of all children with one of the aforementioned diagnoses. A crosscheck was also performed in the departments of neurosurgery, to ensure that children with relevant diagnoses had not been admitted directly without being registered in the regional pediatric department.

Relevant information was collected from the records, made anonymous, and recorded in a spreadsheet. This information included gender, type of symptom and age at symptom onset, age at the first contact with the hospital, clinical findings from the first examination in the hospital, and diagnostic codes. The identity of each patient was kept long enough to ensure that each patient was entered into the study only once. The project was approved by the Regional Ethics Committee, the National Data Inspectorate, and the Directorate for Health and Social Affairs.

RESULTS

Increased HC in Routine Measurements

A total of 138 patients (46%) were referred because of increased HC; 109 (79%) had this as their only symptom, whereas 29 (21%) had additional symptoms. The 109 patients with increased HC as the only symptom had the following diagnoses: hydrocephalus, 99 patients (91%); tumor, 1 patient (1%); hemorrhage, 1 patient (1%); cysts, 8 patients (7%).

Figure 1 shows the distribution of age at the time of diagnosis for patients referred because of increased HC and for those referred for other reasons (all diagnoses included). Increased HC seemed to be the most important symptom of intracranial expansion during the first 10 months of life (Fig 1). Of the patients for whom we had information about the exact age at diagnosis, 123 patients received diagnoses because of increased HC during their first 10 months. Only 7 patients received diagnoses because of increased HC after this age.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Proportions of patients referred because of increased HC, compared with other causes of referral, according to age at symptom onset (n = 285; accurate numbers are provided in the columns). All diagnoses were included. Patients with >1 cause of referral were classified according to the most important/decisive cause. Information on the exact age at symptom onset was lacking for 13 patients.

During the first 12 months of life, 110 (76%) of 145 children diagnosed as having hydrocephalus and only 13 (21%) of 63 children with other diagnoses had increased HC as a cause of referral. The difference between the 2 groups was statistically significant (χ² = 55.4, df = 1; P < .001), which was also the case for the whole study period (χ² = 106.1, df = 1; P < .001).

There was also a significant difference between cysts on one hand and tumors and hemorrhage on the other, both for the first 12 months of life (χ² = 9.6, df = 1; P < .01) and for the whole study period (χ² = 14.1, df = 1; P < .001). More details on the most common diagnostic groups (hydrocephalus, tumors, intracranial hemorrhage, and cysts) are given below.

Hydrocephalus

A total of 173 patients in this study were diagnosed as having hydrocephalus, including 128 boys (74%) and 45 girls (26%). The ages at the time of diagnosis are presented in Fig 2. The overall prevalence of hydrocephalus in the study period was 0.75 cases per 1000 live births. One hundred twenty-four children with hydrocephalus (72%) were referred because of increased HC. The others had symptoms as shown in Table 3. Of the 162 patients with hydrocephalus for whom we had information about the exact age at diagnosis, 141 (87%) experienced the onset of symptoms before 10 months of age, whereas only 21 patients (13%) received diagnoses after that age (Fig 3).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Distribution of ages at the first in-hospital examination (and approximate times of diagnosis) for the children with hydrocephalus (n = 171; accurate numbers are provided in the columns). The exact age at admittance was not known for 2 patients.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3

Proportions of patients with hydrocephalus referred because of increased HC, compared with other causes of referral, according to age at symptom onset (n = 162; accurate numbers are provided in the columns). Information on the exact age at symptom onset was lacking for 11 patients.

View this table:

• View inline
• View popup

TABLE 3

Causes of Referral

The mean age at symptom onset for the children with hydrocephalus with increased HC as the main symptom was 4.8 months (range: 0–27 months; median: 4.0 months), whereas the mean age at the first in-hospital examination for the same patients was 8.7 months (range: 0–39 months; median: 7.5 months). Therefore, for this group of children with hydrocephalus, there was a considerable delay from symptom onset to the first contact with the hospital (range: 0–30 months; mean: 3.9 months; median: 3.0 months).

The second most common referral cause for children with hydrocephalus was nausea/vomiting (9%). The mean age at symptom onset for this group was 10.0 months (range: 0.5–45 months; median: 8.5 months), and the corresponding mean age at the first in-hospital examination was 10.8 months (range: 0.5–45 months; median: 9.5 months). Therefore, the delay from symptom onset to the first hospital contact was significantly shorter for this subgroup of patients with hydrocephalus (range: 0–5.5 months; mean: 0.8 months; median: 0 months) than for patients who were diagnosed because of increasing HC. HC was also measured for 165 (95%) of the patients with hydrocephalus at the first in-hospital examination; 94 (57%) of those patients had HC of >97.5 percentile.

DISCUSSION

To our knowledge, this is the first population-based study to investigate the role and importance of the widely used routine of measuring HC in young children, although other studies have dealt with HC more generally.⁹ Our results indicate that the symptoms of intracranial expansive conditions may vary considerably from one condition to another and that increased HC as a debut symptom seems important only for infants with hydrocephalus and, to a certain degree, patients with cysts. For intracranial tumors and other expansive conditions, increased HC is very rarely the debut symptom that causes suspicion and leads to diagnosis. For these conditions, other symptoms, such as nausea/vomiting, unsteadiness, drowsiness, neurologic deficits, or headache, seem to be more common as the initial symptoms, observations that correspond well to those of earlier studies.^10–12

Our data also indicate that HC measurements are important only during the first 10 months of life. In Norway, measurements of HC are performed routinely at regular intervals during the first year of life, whereas other guidelines suggest regular measurements for up to 24 months after birth.¹³ On the basis of our findings, the benefits of extending routine measurements beyond the age of 1 year may be questioned.

The prevalence of hydrocephalus was 0.75 cases per 1000 live births. A similar study from Sweden presented a prevalence of 0.49 cases per 1000 live births.¹⁴ The difference is probably attributable to the inclusion criteria; the Swedish study included only children in need of surgical treatment, whereas we did not distinguish those patient groups.

The present study has the advantage of being based on data drawn from an entire national population, however small, over a period of 4 years. Therefore, one might think that naturally occurring variations in incidence are compensated for and that our results reflect the importance and limitations of routine HC measurements in infants. On the other hand, our study has the methodologic weaknesses associated with all retrospective studies. A prospective study, based on HC measurements obtained for all children in the cohort, would have given us additional information (eg, about how often an apparently abnormal increase in HC was not caused by any pathologic condition); however, that was beyond the scope and the means of the present study. HC measurements also are important in detecting conditions involving slower-than-normal head growth (eg, craniosynostosis or microcephali).¹⁵ The study focused on intracranial expansive conditions, however.

CONCLUSIONS

This study indicates that routine measurements of HC are of value mainly for the early detection of hydrocephalic conditions and, to a certain degree, intracranial cysts during the first 10 months of life. Intracranial tumors and hemorrhage yield other, often more-dramatic, debut symptoms.