Vaccinating Welders against Pneumococcus: Evidence from a Systematic Review and Meta-Analysis

Abstract

Workers occupationally exposed to welding dusts and fumes have been suspected to be at increased risk of invasive pneumococcal disease (IPD). Since the 2010s, the United Kingdom Department of Health and the German Ständige Impfkommission (STIKO) actively recommend welders undergo immunization with the 23-valent polysaccharide (PPV23) pneumococcal vaccine, but this recommendation has not been extensively shared by international health authorities. The present meta-analysis was therefore designed to collect available evidence on the occurrence of pneumococcal infection and IPD among welders and workers exposed to welding fumes, in order to ascertain the effective base of evidence for this recommendation. PubMed, Embase and MedRxiv databases were searched without a timeframe restriction for the occurrence of pneumococcal infections and IPD among welders and workers exposed to metal dusts, and articles meeting the inclusion criteria were included in a random-effect meta-analysis model. From 854 entries, 14 articles (1.6%) underwent quantitative analysis, including eight retrospective studies (publication range: 1980-2010), and six reports of professional clusters in shipbuilding (range: 2017-2020). Welders had an increased likelihood of developing IPD compared with non-welders (odds ratio 2.59, 95% CI 2.00-3.35, I² = 0%, p = 0.58), and an increased likelihood of dying from IPD (standardized mortality ratio (SMR) 2.42, 95% CI 1.96-2.99, I² = 0%, p = 0.58). Serotype typing was available for 72 cases, 60.3% of which were represented by serotype 4, followed by 12F (19.2%) and serotype 8 (8.2%). Although the available data derive from a limited number of studies, available results suggest that pneumococcal vaccination should be recommended for workers exposed to welding fumes, and vaccination strategies should consider the delivery of recombinant formulates in order to combine the direct protection against serotypes of occupational interest with the mucosal immunization, reducing the circulation of the pathogen in occupational settings characterized by close interpersonal contact.

Keywords: Streptococcus pneumoniae; invasive pneumococcal disease; occupational disease; pneumococcus; prevalence.

Figure A1

Summary of risk of bias assessment for case series studies according to the National Toxicology Program (NTP)’s Office of Health Assessment and Translation (OHAT) handbook and respective risk of bias (ROB) tool [23,24].

Figure A2

Summary of risk of bias assessment for observational studies according to the National Toxicology Program (NTP)’s Office of Health Assessment and Translation (OHAT) handbook and respective risk of bias (ROB) tool [23,24].

Figure A3

Funnel plots of the studies included in the quantitative analysis: (a) notification rate; (b) hospitalization rate; (c) admission to ICU; (d) death rate; (e) estimates on attack rate; (f) standardized mortality ratio (SMR); (g) estimates on odds ratio.

Figure A4

Radial plots for the studies included in the quantitative analysis: (a) notification rate; (b) hospitalization rate; (c) admission to ICU; (d) death rate; (e) estimates on attack rate; (f) standardized mortality ratio (SMR); (g) estimates on odds ratio.

Figure 1

Flowchart of included studies.

Figure 2

Timeframe of the collected studies [8,9,13,29,33,34,35,47].

Figure 3

Forest plot for notification rate (a), hospitalization rates (b), admission to Intensive Care Unit (ICU) (c), and deaths (d) as collected from 5 out of 6 case series included in the present study [30,32,37,38,42].

Figure 3

Forest plot for notification rate (a), hospitalization rates (b), admission to Intensive Care Unit (ICU) (c), and deaths (d) as collected from 5 out of 6 case series included in the present study [30,32,37,38,42].

Figure 4

Forest plot of pooled attack rate for pneumococcal pneumonia (PNX) and invasive pneumococcal diseases in welders. Overall, an attack rate equal to 4.22 per 100 person-years, 95% CI 0.86 to 7.58, was calculated through a random effect model, with a very high heterogeneity (I² = 92%) [30,32,37,38,42].

Figure 5

Forest plot of pooled standardized mortality ratio (SMR) for pneumococcal pneumonia (PNX) and invasive pneumococcal diseases in welders. Overall, a pooled SMR equal to 2.42, 95% CI 1.96 to 2.99, was calculated, with a very low heterogeneity (I² = 0%) [29,33,35].

Figure 6

Forest plot of pooled odds ratio (ORs) for invasive pneumococcal diseases (IPD) in welders compared to cases occurring in non-welders. Overall, a pooled OR equal to 2.59, 95% CI 2.00 to 3.35, was calculated, with a very low heterogeneity (I² = 0%) [8,13,34].