Immunogenicity of sequential pneumococcal vaccination in subjects splenectomised for hereditary spherocytosis


Markus A. Rose, Children's Hospital, Goethe University, Theodor Stern Kai 7, 60590 Frankfurt, Germany.


Splenectomy predisposes for invasive pneumococcal disease. We investigated the immune response of splenectomised hereditary spherocytosis (HS) patients upon sequential pneumococcal vaccination. Thirty-nine HS-patients (2- to 18-year-old) had undergone near-total or total splenectomy. All received one dose of 7-valent pneumococcal conjugate vaccine (PCV-7) and 23-valent-pneumococcal-polysaccharide vaccine (PPV-23) 2 months apart. Pneumococcal antibodies against serotypes 5/6B/7/14/18C/19F/23F and immunoglobulin serum concentrations were determined before PCV-7 and 4 weeks after PPV-23. Significant rises in antibody geometric mean concentrations were observed after PCV-7 except for serotypes 5 and 7, which increased after PPV-23. We found no impact of the mode of splenectomy.

Hereditary spherocytosis (HS) is the most frequent haemolytic anaemia in Europe. Early splenectomy is often required to prevent complications (Eber et al, 1990). As the risk for overwhelming postsplenectomy infection (OPSI) remains life long (Eber et al, 1999), pneumococcal immunisation is recommended. We assessed serum pneumococcal antibody concentrations from splenectomised individuals before and after 7-valent pneumococcal conjugate vaccine (PCV-7) and after subsequent 23-valent-pneumococcal-polysaccharide vaccine (PPV-23) (sequential pneumococcal vaccination; SPV). The incidence of infections after splenectomy because of haematological diseases seems to be higher after trauma, as intraperitoneal spleen fragments can preserve the lienal function (Schwartz, 1996). Thus, increasing numbers of centres perform near-total splenectomy, conserving a 10 cm3 piece of spleen to maintain phagocytosis (Stoehr et al, 2005). The present study also attempted to evaluate the impact of the mode of splenectomy on the immunogenicity of SPV.


This prospective open multi-centre study was performed between September 2002 and August 2003. Inclusion criteria were a history of splenectomy between April 1982 and August 2003 when aged between 2 and 18 years. Historically, patients underwent mainly complete splenectomy, while in more recent years the near-total procedure has been favoured. Exclusion criteria were other severe medical conditions, other vaccinations within the 2 months prior to study, or an intolerance of vaccine ingredients. Subjects underwent 99 m Tc-whole body scan to detect intact spleen tissue. The intravenous application was equivalent to 1·85 MBq/kg bodyweight, corresponding to an effective dose of 1·3–1·8 mSv. Written informed consent was obtained from all participants. The study was approved by the local institutional ethics committee and performed in accordance with the declarations of Helsinki (1964), Edinburgh (2000), and national radiation protection regulations. A detailed description of the applied vaccines, serological analysis, and criteria predicting protection from invasive pneumococcal disease was given in Rose et al (2005). We calculated the geometric mean concentrations (GMCs) with 95% confidence intervals (CI) and the proportion of patients with antibody concentrations ≥1·0 μg/ml, which is considered to be a protective level, for all serotypes pre- and post-vaccination. Values were compared using Wilson's test. The U-test (Mann–Whitney) was performed for between-group comparison. The Statistical Package for the Social Sciences (spss) for Windows® software, version 11.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Probability (P) values <0·05 were considered to be statistically significant.

Results and discussion

Hereditary spherocytosis patients were stratified as ‘completely-’ (group A) or ‘near-total splenectomised’ (group B, Table I). Healthy post-traumatic splenectomised individuals served as controls (group C). Immunoglobulin serum concentrations were similar in the different groups. For group A, median IgG was 11·2 g/l (range 6·8–18·9) compared with 10·4 (5·5–15·9) in group B and 12·0 (11·6–12·3) for group C. Median IgA was 2·3 g/l (0·34–4·8) in group A compared with 1·6 (1·0–2·6) in group B and 1·9 (1·7–2·0) in group C. Median IgM was 0·78 g/l (0·32–2·4) in group A vs. 0·76 g/l (0·33–1·6) in group B and 0·85 (0·73–0·97) in group C. Initial GMCs of specific pneumococcal antibodies were similar for all groups. Fig 1 illustrates values for groups A and B. Although all subjects had been immunised with PPV-23 before splenectomy, an alarming high number of splenectomised individuals did not reach the protective level (depending on the serotype up to 46·3%). Eight weeks after PCV-7, all subjects showed significant increases of their serum GMCs for all PCV-7 contained serotypes. Only a few patients did not reach the protective level, depending on the serotype 2·4–14·6%. Four weeks after PPV-23, specific pneumococcal antibodies for serotypes 5 and 7 also increased significantly from initial values and nearly all subjects achieved the protective level. For the other serotypes, no further significant increase was observed except for serotype 19F in group B. There were no significant differences between HS-patients and group C controls at any time.

Table I.  Features of study population.
 Group A (n = 16)Group B (n = 23)Group C (n = 3)
  1. Group A, completely splenectomised HS-patients; Group B, near total splenectomised HS-patients; Group C, healthy controls splenectomised after trauma; HS, hereditary spherocytosis; f, female; m, male.

  2. Ages and times (years); medians and 95% CI are shown.

  3. *Group C subjects were older than in group A (P < 0·05) and group B (P = 0·052).

  4. † Group B subjects were younger than group A (P < 0·01) and group C (P = 0·06).

  5. ‡Time was longer in group A than group B (P < 0·001). PCV-7, 7-valent pneumococcal conjugate vaccine; PPV-23, 23-valent pneumococcal polysaccharide vaccine.

  6. §Data only available for one subject.

Distribution of gender (f:m)10:613:10 2:1
Age at splenectomy*9·9 (3·3–22·0)9·6 (4·9–22·7)18·6 (12·3–24·2)
Age at study entrance†18·2 (7·4–26·7)13·4 (5·3–24·6)23·3 (14·7–25·3)
Time between last PPV-23 and PCV-7‡4·3 (0·9–16·5)2·3 (0·8–6·7) 0·67§
Time between splenectomy and PCV-7‡6·5 (0·8–20·7)0·9 (0·2–7·9) 3·2 (2·3–6·1)
Figure 1.

Pneumococcal serum antibody geometric mean concentrations (95% CI) for different serotypes in hereditary spherocytosis patients following sequential immunisation. Group A, totally-, group B, near-totally splenectomised. No significant differences were found except after pneumococcal conjugate vaccine-7 for 19F and 23F (P < 0·05, each) and after pneumococcal polysaccharide vaccine-23 for 23F (P < 0·05).

Children who undergo splenectomy at young age are at an even higher risk of OPSI (Schwartz, 1996). In the absence of functional splenic tissue, the ability to mount a primary antibody response to pneumococcal polysaccharide vaccine is also missing. This is in accordance with our observation of an alarming weak immunogenicity of the antecedent PPV-23. Also in a study on 76 adult patients splenectomised for haematological disorders, 28% showed a poor response upon PPV-23, resulting in five episodes of pneumococcal infections despite repeated re-vaccination (Cherif et al, 2005). Conjugation of capsular polysaccharides to a protein carrier converts the nature of the antipolysaccharide responses from T-cell independent to T-cell dependent, which aids secondary lymphoid tissues in compensating for the absence of a spleen (Breukels et al, 2001). The poor response to bacterial polysaccharides because of an absence of IgM memory B cells is bypassed, and T-helper cells stimulate polysaccharide-specific B cells to mature into antibody-producing plasma cells or into memory cells (Siber, 1994). Nonetheless, the protection provided is limited to the antigens present in the vaccine. Although higher-valent PCVs are under evaluation, they do not have the same coverage of serotypes as PPV-23. Thus, and based on a study on children with sickle cell haemoglobinopathy, SPV has been recommended for patients at an increased risk for pneumococcal infections (Vernacchio et al, 1998; American Academy of Pediatrics. Committee on Infectious Diseases, 2003). The time between PCV-7 and PPV-23 immunisation seems to have an impact on the antibody response. In those patients with sickle cell haemoglobinopathy and in contrast to our findings, a further increase in PCV-7 included serotypes was measured when PPV-23 was given at least 12 months after PCV-7. Also the time since splenectomy may have an impact in those splenectomised <10 years before PCV-7 immunisation, generating lower serum antibody titres. Individuals who had recently been splenectomised had depleted circulating memory B cells, but the switched memory B cells, which may be responsible for the response to PCV-7, were replaced over time (Kruetzmann et al, 2003). Accordingly, we found a trend of lower specific pneumococcal GMCs in group B, which had a significantly shorter period of time between splenectomy and PCV-7 immunisation.

In conclusion, all our splenectomised individuals achieved titres of functional antibodies greater than the protective threshold, implying that SPV induces sufficient immune responses in HS-patients and irrespective of the mode of splenectomy. Nevertheless, even long-term penicillin-prophylaxis and pneumococcal immunisation produce an incomplete protection towards encapsulated bacteria. Thus, spleen-saving procedures have the potential to preserve at least a part of the splenic immune function and should be pursued especially in immunologically immature or immunodepressed patients.

Financial support


Conflict of interest

None. Several of the authors have declared a financial interest in a company whose product was studied in the present work. G.A.S. has once received honoraria from Wyeth for a translation. MA.R. and S.Z. have occasionally received honoraria for lectures and attendance at paediatric advisory boards from Wyeth and Merck Sharpe and Dohme.