| Summary: | Dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science in Medicine.
Johannesburg, 2015 === Pertussis, caused by Bordetella pertussis, is a vaccine-preventable disease affecting persons of all ages. Despite vaccination with either the whole-cell or acellular vaccine, the burden of pertussis has increased worldwide. The acellular vaccine was licensed in South Africa in 2009, replacing the whole-cell vaccine; however, due to no active surveillance, pertussis is underestimated in this country. This study describes the burden of disease caused by B. pertussis and other Bordetella species in patients with severe respiratory illness (SRI), influenza-like illness (ILI) and controls.
Prospective, active surveillance was conducted amongst SRI and ILI patients and controls at two sentinel sites in South Africa. Patients who met the case definitions were enrolled from May 2012 to October 2014. Clinical and demographic data were collected. Induced sputum was collected from SRI patients only and combined nasopharyngeal/oropharyngeal specimens were collected from all patients and controls. Real-time polymerase chain reaction (PCR) was used to target the insertion sequences IS481, pIS1001, hIS1001 and pertussis toxin gene ptxS1. All data were analysed in Microsoft Excel (Microsoft Corporation). Statistical significance was determined using the chi-squared test and univariate logistic regression at p <0.05 for all parameters.
Of 8569 cases that were enrolled and tested, 118 [1.4%, 118/8569 (95% CI 1.1 – 1.6)] were positive for B. pertussis of which 2% [80/3982 (95% CI 1.6 – 2.5)] presented with SRI, 1% [32/3243 (95% CI 0.7 – 1.4)] with ILI and 0.4% [6/1344 (95% CI 0.2 – 1.0)] were asymptomatic. Positive cases were stratified into confirmed pertussis and probable pertussis based on cycle threshold (Ct) value cut-offs generated by real-time PCR for IS481. Within the SRI population, there were more probable than confirmed pertussis cases [51/3982, 1.3%
vs. 29/3982, 0.7%; p=0.02] and within the ILI group there were 0.5% confirmed and probable cases, respectively [15/3243, 0.5% vs. 17/3243, 0.5%; p=0.86]. The highest detection rate of pertussis in SRI positive cases was in the ≥65 year olds (2.8%, 6/208) and for the ILI positive cases the highest detection rate was in the 1-4 year olds (1.5%, 9/614). Pertussis disease was observed mainly in the winter and spring months with a 15% increase in disease detected in August 2014. The B. pertussis attributable fraction was 67% (95% confidence interval [CI] 18.49 – 86.63) for SRI positive cases. Fifty-eight percent (46/80) of B. pertussis positive cases were co-infected with respiratory bacteria (Streptococcus pneumoniae, Haemophilus influenzae, Legionella spp. or Mycoplasma pneumoniae) or viruses (influenza, respiratory syncytial virus (RSV), human metapneumovirus or other viruses (adenovirus, enterovirus, parainfluenza or rhinovirus). HIV status and full pertussis vaccination for age did not affect B. pertussis positivity.
B. parapertussis was detected in 1% [40/3982 (95% CI 0.7 – 1.4)] of the SRI population, 0.6% [18/3243 (95% CI 0.3 – 0.9)] of the ILI population and in 0.1% [2/1344 (0.02 – 0.5)] of asymptomatic individuals. The highest detection rate for the SRI (1.6%, 8/497) and ILI (1.5%, 9/614) positive cases were in the 1-4 year olds. The B. parapertussis attributable fraction was 80% (95% confidence interval [CI] 12.52 – 95.38) for SRI cases. Four cases tested positive for B. bronchiseptica, of which one individual was HIV positive.
B. pertussis, B. parapertussis and B. bronchiseptica were detected despite the case definitions not being ideal for the detection of these pathogens. Bordetella spp. was detected in all age groups tested. This study generates baseline data for pertussis in South Africa and surveillance is ongoing.
|
|---|
