METHODS

Subjects and immunisations

Preterm infants born at less than 32 weeks gestation were recruited from five neonatal units in one UK region. The study had ethical approval from the four local research ethics committees serving the five centres, and informed consent was obtained from the parents before enrolment. All infants received three doses of a combined diphtheria/tetanus/acellular pertussis (DTaP)-Hib conjugate vaccine (Infanrix-Hib; GlaxoSmithKline, Uxbridge, Middlesex, UK; 0.5 ml dose containing 30 IU diphtheria toxoid, 40 IU tetanus toxoid, 25 μg pertussis toxin, 25 μg filamentous haemagglutinin, 8 μg pertactin, and 10 μg purified capsular polysaccharide of Hib covalently bound to about 30 μg tetanus toxoid) and meningococcal serogroup C conjugate vaccine (Meningitec; Wyeth, Collegeville, Pennsylvania, USA; 0.5 ml dose containing 10 μg meningococcal serogroup C oligosaccharide conjugated to about 15 μg CRM₁₉₇ protein) at 2, 3, and 4 months of age.³ Preterm infants with a Hib IgG < 1.0 μg/ml after the three primary immunisations were offered a fourth dose of Hib conjugate vaccine (Act-HIB; Aventis Pasteur MSD, Lyon, France; 0.5 ml dose containing 10 μg purified Hib capsular polysaccharide covalently bound to 24 μg tetanus protein).

Assays

Blood was taken by needle venepuncture into Microtainer Serum Separator tubes four weeks after the additional fourth Hib vaccine was given. Antibody concentrations to Hib were assayed at the Centre for Applied Microbiology and Research (CAMR), Salisbury, Wiltshire, UK by enzyme linked immunosorbent assay using a standardised protocol as previously described.⁴ The lower limit of detection of the assay was 0.15 μg/ml. Samples with undetectable Hib IgG were assigned a value of half the lower limit of detection (0.08 μg/ml). Hib IgG avidity was determined by the Immunobiology Unit, Institute of Child Health, London, UK as previously described.⁵ Avidity results in this study are multiplied by 100 compared with previous reports for computational ease. Avidity could only be measured where Hib IgG was > 0.5 μg/ml. The lower limit of detection of the avidity assay was 50 avidity units. Sera with unrecordable avidity were assigned a value equal to half the lower limit of detection (25 avidity units).

Statistical analysis

Hib IgG concentrations after the third and fourth doses were log₁₀ transformed to achieve normality. Hib IgG geometric mean concentrations (GMCs) and avidity index (GMAI) were calculated with 95% confidence intervals (95% CI) and compared using Student’s t test. The effect of patient variables on antibody responses was assessed using linear regression. Variables with p < 0.2 were entered into a multivariable regression model.

RESULTS

A total of 122 preterm infants (mean gestational age at birth 29.1 weeks (range 24.2–31.9); mean birth weight 1266 g (range 490–2100)) were immunised at 2, 3, and 4 months of age. The Hib IgG response in 105 of these infants after primary immunisations has been reported previously.³ The Hib IgG GMC after the primary immunisation for all the preterm infants was 0.29 μg/ml. Ninety six (79%) had Hib IgG < 1.0 μg/ml after primary immunisations and received a fourth immunisation; in 51 (42%) Hib IgG was < 0.15 μg/ml (historical level for short term protection). The mean age at receipt of the additional dose was 7.8 months (range 6.2–11.5).

In these 96 infants, the Hib IgG GMC after the primary immunisation was 0.17 μg/ml (95% CI 0.14 to 0.20). This rose to 4.68 μg/ml (95% CI 3.36 to 6.57) after the fourth dose (p < 0.0001). Table 1 shows GMCs for infants with Hib IgG concentration < 0.15 μg/ml and ⩾ 0.15 μg/ml after the primary and booster immunisations. Table 2 shows the proportions of infants achieving Hib IgG concentrations ⩾ 0.15 and 1.0 μg/ml after the primary and additional fourth immunisations. Five infants failed to achieve a Hib IgG ⩾ 0.15 μg/ml after the fourth dose. There was no difference between these infants and those who did achieve concentrations ⩾ 0.15 μg/ml in mean age at receipt of the fourth dose (8.1 and 7.8 months respectively, p  =  0.52).

The effect of a number of patient variables (gestational age at birth, birth weight, age at third dose, Hib IgG after third dose, age at fourth dose, number of doses of antenatal steroids, and number of courses of postnatal steroids) on IgG response to the fourth dose were assessed using linear regression. Variables with p < 0.2 in linear regression (age at third dose, p  =  0.063; IgG after third dose, p < 0.001; age at fourth dose, p  =  0.034) were analysed further using multivariable regression. Table 3 shows the results of the multivariable regression analysis. Of note, IgG response to a fourth dose correlated positively with age at fourth dose in linear regression, but this ceased to be significant when controlled for age at completion of the primary immunisations.

The Hib GMAI was determined after the third dose, and after the fourth dose when given. As the response to the primary immunisations was so low in this group, it was only possible to measure GMAI after the third dose in 11 infants. GMAI after the third dose in infants not requiring an additional dose was also measured. Table 4 shows the results. Table 1 shows Hib GMAIs after the fourth dose for infants with Hib IgG < 0.15 μg/ml or ⩾ 0.15 μg/ml after the primary immunisations.

Of those infants who were given an additional dose of Hib, 10 had received at least one course of postnatal steroids for chronic lung disease. The Hib GMC for these infants after the primary immunisations was 0.16 μg/ml. This rose to 3.68 μg/ml after the fourth dose (p  =  0.002).

DISCUSSION

We have shown that preterm infants with very low IgG responses to Hib conjugate vaccine mount an adequate response to an early additional fourth dose of Hib. The GMC after the booster, even in infants who were < 0.15 μg/ml after the primary immunisations, compares favourably with the concentrations achieved after the three dose primary course in term infants receiving a separate Hib vaccine at 2/3/4 months (Hib IgG GMC 4.50 μg/ml, 95% CI 3.21 to 6.32).⁶ These findings, in preterm infants with IgG concentrations after the primary immunisations lower than those of term infants³ indicate that good responses can be expected in term infants boosted soon after primary immunisation, as occurred recently in the United Kingdom.

We are unable to say with certainty whether the response to the additional dose represents a primary or a memory response, as there are no comparable data on the response expected in naïve and primed infants to a booster dose given at this early age. However, as the GMAI after the booster was higher in the group with Hib IgG concentrations ⩾ 0.15 μg/ml after the primary immunisations than in those with concentrations of < 0.15 μg/ml, this suggests some degree of memory in the former group. The GMAI after the booster in those with concentrations < 0.15 μg/ml was similar to that seen in the 26 infants with concentrations >1.0 μg/ml, suggesting a primary rather than a memory response.

This study strongly suggests that all infants will benefit from an additional fourth Hib dose, irrespective of the timing of its delivery or the antibody concentrations achieved after the third dose. For children in whom the fourth dose may be a priming rather than a boosting dose, there is the question of whether a subsequent booster would be needed. We intend to follow up this cohort of preterm infants to compare antibody persistence in those for whom the fourth dose was acting as a booster compared with those in whom it was probably a priming immunisation.