Fifty-three fatal cases in a recent Californian outbreak were significantly more likely to have lower birth weight, younger gestational age, younger age at time of cough onset, and higher peak white blood cell (WBC) and lymphocyte counts. They were less likely to have received macrolide antibiotics but more likely to have received steroids or nitric oxide and to develop pulmonary hypertension, seizures, encephalitis, and pneumonia. In multivariate analyses, peak WBC count, birth weight, intubation, and receipt of nitric oxide were predictors of death. 26082502
In the same epidemic, patients with pertussis were more often afebrile, had more visits before admission, and had longer hospital stays compared to those cases with viral respiratory infection (RSV and influenza). 21925678
Most pertussis deaths occur in infants, particularly in those < 3 months of age.
The specific host risk factors then, for infant pertussis mortality, relate to immaturity: immaturity of the cardiorespiratory system (particularly the airways and lungs); and immaturity of the immune system.
Risk factor for Infant Pertussis Mortality 19209089
- birth weight < 2,500 g
- female sex
- Apgar score < 8
- mother with < 12 years education
Pertussis is also associated with encephalopathy and seizure in infants. A Danish study showed the risk of developing subsequent epilepsy was increased in children with hospital-diagnosed pertussis infections, albeit with a low absolute risk (Hazard Ratio 1.7). 26529162
Fulminant refractory cardiorespiratory failure was the ultimate cause of death of ten patients under 8 weeks of age in a Birmingham study, with pathological evidence of necrotising bronchitis and bronchiolitis with extensive areas of necrosis of the alveolar epithelium, including hyaline membrane disease in those cases with viral co-infection. The pulmonary vasculature was filled with an inflammatory infiltrate and the thymus, spleen, and lymph nodes showed immunodepletion. Co-infections isolated were Parainfluenza type 3, Moraxella catarrhalis, RSV, MRSA and MSSA, H. influenzae, and Candida tropicalis. These features were likely a combination of the effects of pertussis toxin, tenacious airway secretions, overwhelming inflammatory response, leukostasis, cellular apoptosis, and secondary immunocompromise. 19725100
Severe complications of pertussis include pneumonia, encephalopathy and meningoencephalitis. In addition, infants may experience weight loss, bronchitis, otitis media, apnoea, cyanosis, inguinal hernia and rectal prolapse. 15876922
The classical features of paroxysmal pertussis are often absent in older children and adults and after vaccination. Paroxysms (34%), post-tussive emesis (30%), and wheezing on physical examination (8%) were associated with a longer duration of symptoms at diagnosis. Wheezing was associated with a delay in diagnosis, while those with a documented exposure were associated with a more timely pertussis diagnosis. 24168977
In partially immunised children, it may present as a persistent cough > 7 days (and up to 3-4 months in some trials), although this may also be due to other infections — commonly Adenovirus, Parainfluenza viruses 1, 2, and 3, Mycoplasma pneumoniae, and Respiratory Syncytial virus. 9686733
Cough duration is longer in children than in adults with pertussis (median cough duration 112 days versus 42 days); individuals may take even longer to recover fully and regain previous levels of exercise tolerance. 21034844
The duration of cough can be greatly influenced by vaccination status. 14595048
Congestion and cough can pre-date the paroxysms by up to 1 week in a third of cases, whereas two-thirds of children under 5 will develop a paroxysmal cough within the first week of their illness. And for 9 out of 10 children in this age group, paroxysmal cough will persist for more than 21 days. 8584314
In children (and adults) with persistent cough, paroxysmal coughing is the most sensitive indicator of pertussis, but has poor specificity and limited diagnostic value. Vomiting and whooping, particularly in combination, are strong predictors of pertussis. 21034844
So what to do when seeing patients, in a community setting, with a relatively mild illness in whom you consider the possibility of pertussis or where the possibility is raised by the patient or family members?
The burden of even mild pertussis is not insignificant, with a disease duration of more than 3 weeks, considerable sleep disturbance, and isolation. 1528646
A diagnosis of pertussis may be confirmed by culture, Polymerase Chain Reaction (PCR) or serology. Single estimates of anti-pertussis toxin (PT) antibody titres in blood or oral fluid samples are highly specific. 21034844
In a German trial, 26 of 28 B. pertussis culture-positive cases also had positive PCR as did 82 additional (i.e. culture-negative) cases, demonstrating a 4-fold increase in identification by PCR compared to culture. PCR identifies many milder illnesses not identified by culture. 7761186
There are currently no proven efficacious treatments for pertussis-induced cough. Treatment with macrolide antibiotics reduces the duration of an individual’s infectious period, but does not alter the duration of cough. 21034844
Sudden Infant Death mortality rate followed significantly the monthly prevalence of pertussis in Stockholm and all of Sweden (where an immunisation programme against Bordetella pertussis was ceased in 1979). And when controlling for seasonality, a significant correlation persisted in the urban area of Stockholm. 9177988
Case management
- Antibiotic choice and dose depend on patient age
- Neonate: azithromycin 10 mg/kg OD x 5 days
- 2-6 Months: azithromycin 10 mg/kg OD x 5 days
- > 6 Months: azithromycin 10 mg/kg (max 500 mg) as single dose Day 1, then 5 mg/kg (max 250 mg) OD for Days 2 to 5
- Isolate index case: non-infectious after 5 days treatment or 21 days from onset of cough
- Test and notify
Australian guidelines for the public health management of pertussis recommend antibiotic treatment of the index case with exclusion from childcare, school, work or other environments where high-risk contacts may be present, until they are non-infectious (that is, after five days of antibiotic treatment).
Contact Management
- Antibiotic prophylaxis – for unimmunised infants, late pregnancy, an infectious source
- Vaccinate
Australian guidelines recommend a single dose of an acellular vaccine for contacts of pertussis older than eight years, and catch-up vaccination for unvaccinated or partially vaccinated (incomplete infant vaccination) contacts up to their eighth birthday.
Macrolide antibiotics are the preferred drug for both treatment and prophylaxis. Azithromycin is the antibiotic of choice in neonates.
Erythromycin should only be used in neonates when azithromycin is unavailable, because of a risk of pyloric hypertrophy.
Clarithromycin is not recommended in neonates but can be used at a dose of 7.5 mg/kg BD for 7 days in other infants and older children.
For macrolide intolerant patients of more than 2 months of age, trimethoprim-sulfamethoxazole can be used at a dose of 4/20 mg/kg BD for 7 days.
The dosing of erythromycin, too, is age-dependent:
- < 7 Days old: 10 mg/kg BD x 7 days
- 7-28 Days old: 10 mg/kg Q8H x 7 days
- 2-6 Months old: 10 mg/kg Q6H x 7 days
- > 6 Months old: 7.5 mg/kg (max 500 mg/dose) BD x 7 days
SUMMARY
Pertussis is highly infectious (transmitted in up to 50% of casual contacts and 80% of susceptible household contacts) and humans are the only known disease host. Be prepared to treat the younger infant on suspicion. Manage the case but also manage the contacts. Ensure pertussis booster vaccination of adults and adolescents in close contact with an infant, including pre-partum in mums.
Other References
