Introduction Immunisation against pneumococcus has been shown to reduce pneumonia in rheumatoid arthritis (RA). There is concern that methotrexate may reduce its efficacy. There are very few objective data on the effect of methotrexate on the efficacy of pneumococcal vaccination with pneumovax, and no objective evidence on whether revaccination is necessary in RA patients on methotrexate.
Methods The authors collected information from 180 RA patients on methotrexate relating to their vaccination status and assayed their pneumococcal antibody levels. Data on pulmonary infection were retrieved in the same patients over the preceding decade.
Results Full data were available for 152 patients, of whom 28 had never been vaccinated against pneumococcus. Median levels were significantly higher in those who had been vaccinated. Unvaccinated patients and those taking oral prednisone were more likely to have had pneumonia in the previous 10 years. The RR for developing pneumonia among non-vaccinated patients was 9.7 (p=0.005) and among steroid-treated patients was 6.5 (p=0.001), after adjusting for age, gender, disease duration and comorbidity. No significant correlation was found between pneumococcal antibody levels and time since vaccination.
Conclusions This study suggests that a single administration of pneumovax early in RA offers up to 10 years protection against the development of pneumococcal pneumonia in RA patients on methotrexate.
Statistics from Altmetric.com
Life expectancy is reduced in rheumatoid arthritis (RA).1 Excess mortality is partly due to the increased risk of serious infections in patients with RA,2 and the lower respiratory tract appears to be one of the commonest sites of infection.3 Pneumonia is associated with an increase in both morbidity and mortality in RA patients.4 Several factors contribute to these observations, and these include the presence of previous pulmonary disease and the use of long-term oral steroids.4,–,6 Immunisation against pneumococcus has also been proposed as an important mechanism for the prevention of pneumonia,7 8 and we recently showed how the introduction of such an immunisation programme can contribute to a major reduction in the incidence of pneumonia in RA.9 There is concern that the use of certain drugs, for example, methotrexate, may reduce the value of such a vaccination programme,10 11 and this underlines the importance of vaccination early in the disease process. There are as yet few data on the effect of methotrexate on the efficacy of pneumococcal vaccination, and no objective evidence on whether revaccination is necessary in RA patients on methotrexate.
We identified 180 consecutive outpatients who met the 1987 American Rheumatism Association criteria for RA12 and who had been on a stable dose of methotrexate for at least 12 months. We collected information from each patient on the following: dose and duration of methotrexate therapy; dose and duration of oral steroid therapy; whether or not they had received pneumovax at any point over the preceding 10 years and, if so, the date of administration. These data were cross-referenced with primary care records. If details of the administration of pneumovax could not be confirmed, the patient's data were excluded from analysis. Blood was then taken for the assay of pneumococcal antibody levels, which were analysed in batches, blinded to other clinical details, using a standard commercial (MK012)ELISA assay in kit form, which uses 23 valent polysaccharide (pneumovax) as the antigen in microtitre plates to detect polysaccharide-specific antibodies in serum.
We also collected data on pulmonary infection over the preceding decade in all study patients. This was obtained by a study of both primary and secondary care records in all patients, and required radiological evidence of infection for confirmation. All drug therapy at the date of diagnosis of pneumonia was confirmed. Ethics approval for the study was obtained from the regional ethics committee, and the work was undertaken during the spring of 2009.
Statistical advice suggested the number of patients required to show a significant difference in protective levels between vaccinated and non-vaccinated RA patients with 95% probability was 180, allowing for a 20% loss of data. We measured pneumococcal antibody levels in each patient and compared these in those who had been vaccinated with those who had not, using Student's unpaired t test. Absolute antibody levels under 35 units were considered to indicate unprotected status, based on levels affording protection in splenectomised patients.13 We compared variables between vaccinated and unvaccinated groups, and those with and without previous pneumonia, using Fisher's exact test. To compare antibody levels with both the dose and duration of methotrexate therapy, we calculated Spearman's correlation coefficient. Within the vaccinated group, we also correlated antibody levels with time since vaccination. We calculated RR of pneumonia with 95% CI and the corresponding z statistics for both non-immunised and steroid-treated patients.
Our 180 consecutive RA patients had a mean disease duration of 13 years and were representative of our RA population as a whole, with 68% rheumatoid factor positivity and erosions evident radiologically in 42% at 5 years. Data on vaccination status and/or date were unconfirmed or unreliable in 28 individuals, and these were excluded from further analysis. Full data were available for the remaining 152 patients, and pneumococcal levels are compared between vaccinated (n=124) and non-vaccinated patients (n=28) (table 1). Levels related to time since vaccination are shown in figure 1. All patients were on methotrexate at the time of vaccination.
No correlation was found between pneumococcal antibody levels and methotrexate dose (r=−0.01), duration (r=−0.03) or time since vaccination (r=−0.07). However, there was a trend for levels to fall from 7 years after vaccination. Methotrexate doses were comparable in vaccinated and non-vaccinated patients, as shown in table 1.
Eleven patients had had pulmonary infection in the preceding 10 years. Seven of these patients had never received pneumovax, and one patient had been vaccinated after pneumonia. The characteristics of patients with pneumonia are compared with those without in table 2. After adjusting for age, gender, disease duration and comorbidity, the RR for developing pneumonia among non-vaccinated patients was 9.7 (3.1–38.7, z statistic 3.825, p=0.005) and among steroid-treated patients was 6.5 (2.5–19.7, z statistic 3.511, p=0.001).
Recent data confirm that pneumovax is effective in reducing pneumonia in older people.14 An accompanying editorial agreed that it should be used to reduce pneumonia in patients with chronic disease, but lamented that evidence of efficacy was still limited.15 Our data show that vaccination of RA patients on methotrexate with pneumovax leads to an approximate doubling of median antibody levels and a halving of the percentage of patients whose antibody levels fall below the protective threshold of 35 units. Does this translate into a subsequent reduction in the incidence of pneumonia in the RA population on methotrexate? Our data suggest that it does. Patients with RA who had received pneumovax were 10 times less likely to develop pneumonia over a 10-year period than those who had not been vaccinated. Furthermore, this mirrors data in the population at large. In a study of 3415 patients with community-acquired pneumonia requiring hospitalisation, individuals who had been vaccinated against pneumococcus previously had a 40% lower rate of mortality or intensive care admission compared with those who had not been vaccinated.16
The potential for certain drugs such as steroids, methotrexate and anti-tumour necrosis factor therapy to reduce the efficacy of the vaccination programme has been raised.10 11 17 However, using patients on methotrexate as a comparator group, a large review of serious infections in RA patients showed no significant increase in pneumonia in patients on anti-tumour necrosis factor therapy. It did demonstrate a dose-related rise in the risk of hospitalisation from pneumonia for those on oral steroids, with RR rising from 2.3 with low-dose therapy to over 4.3 in those on at least 7.5 mg a day of prednisone.18 These data are consistent with other large recent surveys of patients with RA.6 The present data support these observations and reinforce the need for caution when considering the long-term use of oral steroids in patients with RA.
The evidence suggests that methotrexate is not responsible for the increased risk of infections in RA,4 and that patients on methotrexate are no more likely to get pneumonia than those on other disease-modifying therapy.17 18 Indeed, the present study provides further reassurance in that neither the dose nor the duration of methotrexate therapy related to the levels of pneumococcal antibody over the 10 years the survey covered. Our data show that the benefit of vaccination lasts for up to 10 years. There is no need for revaccination on present evidence. All patients with RA should therefore receive pneumovax independent of the nature of their therapy.
We have previously shown that vaccination rates improved significantly after a campaign to increase awareness among both patients and primary care providers.7 However, lower rates have been reported elsewhere.8 19 This underlines the importance of highlighting the need for vaccination among all professionals responsible for the care of patients with RA, and indeed among patient groups too. The antibody concentrations to most pneumococcal vaccine antigens remain elevated for at least 5 years in healthy adults. Generally, response to pneumovax in the population at large remains good at 10 years, but vulnerability develops among older people, those with respiratory disease and the immunosuppressed. In a large study, efficacy did not decline at 9 or more years after vaccination.20 The present study also suggests that the administration of pneumovax early in the disease offers up to 10 years protection against the development of pneumococcal pneumonia. Adoption of this approach is supported by a recently published review of available evidence.21
Certain factors are already known to increase the risk of infection and these include the presence of underlying lung disease, male gender and increasing age.1 3 4 Indeed, patients with underlying interstitial lung disease may be at particular risk as they are more often male, usually older people and have limited pulmonary reserve by definition.22 Vaccination should therefore be offered at the first opportunity to such patients, especially as access to a wider range of immunomodulatory agents for the treatment of interstitial lung disease is realised.23
Competing interests None.
Ethics approval This study was conducted with the approval of the Central Office Regional Ethical Committees (COREC).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.