Microcephaly

Microcephaly refers to an abnormally small head defined as a head circumference (HC) of 3 standard deviations (SD) or more below the mean for the gestational age [1–4]. Microcephaly can be present at birth or it may develop in the first few years of life [6].

Prenatally, Chervenak et al. found fetal microcephaly could not be reliably diagnosed or excluded on the basis of a single ultrasound examination if the HC was between 2 SD and 5 SD below the mean [2]. The Society for Maternal-Fetal Medicine (SMFM) recommends that a detailed neurosonographic examination be done on fetuses with an HC greater than 2 standard deviations below the mean to look for findings such as intracranial calcifications, cerebellar hypoplasia, and other abnormalities. The finding of a sloping forehead should increase the index of suspicion for pathological microcephaly. The SMFM also advises that the diagnosis of pathologic microcephaly is considered certain when the fetal HC is equal to or greater than 5 SD below the mean for the gestational age [4]. A follow-up ultrasound in 3 to 4 weeks is recommended for fetuses with an HC more than 2 SD below the mean.

Pathological microcephaly is usually associated with abnormal neurological findings, intellectual disability, and sometimes seizures. Causes of microcephaly include, but are not limited to, a chromosomal abnormality (trisomy or deletion), a gene defect such as Smith-Lemli-Opitz, an infection such as rubella, cytomegalovirus (CMV), toxoplasmosis, or possibly Zika virus, intrauterine fetal stroke, exposure to high doses of radiation, exposure to drugs such as alcohol, aminopterin, or hydantoin, or a maternal condition such as phenylketonuria (PKU).

Microcephaly occurs in about 2 to 12 babies per 10,000 live births in the United States [5].

Predictive performance of different fetal head circumference curves

Historically, the 3 SD cutoff proposed by Chervenak et al., based on the Jeanty head circumference charts, has been highly sensitive for detecting fetuses at risk for microcephaly but has a relatively low positive predictive value (PPV) for congenital microcephaly, with many fetuses with HC < −3 SD ultimately having normal head size at birth [1,2,7,8].

Leibovitz et al. re-evaluated a cohort of fetuses referred with suspected microcephaly and compared three different reference ranges for fetal HC (the conventional Jeanty/Chervenak chart, the Intergrowth-21st standard, and a population-specific Israeli chart) using the same −3 SD threshold. In that series, the PPV for microcephaly at birth was approximately 56–57% with the conventional Jeanty/Chervenak curve, and increased modestly to about 61.5% with the Intergrowth-21st curve and 66.7% with the Israeli reference curve. The differences in PPV among the three curves were not statistically significant, likely because of the relatively small sample size [7,8].

The World Health Organization (WHO) Fetal Growth Charts provide an international, ultrasound-based standard for fetal biometric measurements, including HC, derived from a large, multinational cohort of low-risk pregnancies [9]. A recent cross-sectional study that compared five major HC norms (Jeanty, Hadlock, Intergrowth-21st, WHO, and the NICHD unified standard) using z-scores found that the WHO standard had the best overall statistical fit to observed HC measurements (mean z-score close to 0, SD of z close to 1, favorable distribution and classification properties) and concluded that the WHO curves are likely the most appropriate for diagnosing abnormal fetal HC. In contrast, the Jeanty (Chervenak) reference recommended in earlier SMFM guidance had a much larger SD and was not sensitive for microcephaly screening [10].

That z-score study was not designed to evaluate the positive predictive value of the WHO fetal HC standard for microcephaly at birth, so formal PPV estimates for WHO curves using a −3 SD cutoff are not yet well established. Because the WHO and Intergrowth-21st standards are both prescriptive, international curves derived from healthy populations, their performance for detecting severe restriction of HC (for example, HC < −3 SD) is expected to be broadly similar in comparable clinical settings. However, the PPV for any chart will still depend heavily on the underlying prevalence of microcephaly and on the presence or absence of additional risk factors and structural abnormalities [7–10].

The positive predictive value (PPV) of identifying microcephaly can be significantly improved by considering it alongside other clinical indicators, such as an atypical cranial shape ratio, extremely low estimated fetal weight, brain structural abnormalities, or pertinent familial history. Alternatively, applying a stricter threshold (e.g., HC at or below −4 standard deviations) may also improve diagnostic accuracy. However, these approaches may reduce sensitivity and potentially postpone diagnosis in certain cases [7,8].

*Values for Jeanty/Chervenak, Intergrowth-21st, and the Israeli reference are derived from the cohort studied by Leibovitz et al.; the differences in PPV between the curves were not statistically significant, and results may not generalize to all populations [7,8]. For the WHO curves, formal PPV for microcephaly at birth has not yet been systematically reported.

Head Circumference with Standard Deviations (pdf) Head Circumference with Standard Deviations (jpg)

References