Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Bordetella pertussis and Other Bordetella spp. Infections
Pertussis, commonly known as “whooping cough,” is an acute infectious illness of the respiratory tract causing disease in all age groups but taking its worst toll in unprotected infants and neonates too young to benefit from immunization. It is caused by Bordetella pertussis and, less frequently, by Bordetella parapertussis and Bordetella holmesii . The first epidemic of pertussis was noted in Paris in 1578 and described by Guillaume de Baillou in 1640. Isolation of B. pertussis was reported in 1906 by Bordet and Gengou. By the 1940s and 1950s, whole-cell pertussis vaccines were demonstrated to be efficacious in children. However, because of concerns relating to serious temporally related adverse events with whole-cell vaccines, acellular pertussis vaccines were developed and came into widespread use in the late 1990s.
It is clear that B. pertussis infections cause their most severe illness in unprotected neonates and young infants and that the major source of infection is unrecognized pertussis in the mother or another family member.
Organisms
Bordetella is a gram-negative, pleomorphic, aerobic bacillus that is grouped together on the basis of genotypic characteristics, and species is differentiated by phenotypic characteristics. Most Bordetella species have relatively simple requirements, but B. pertussis is quite fastidious and is inhibited by constituents in common laboratory media, such as fatty acids, metal ions, sulfides, and peroxides.
The genus Bordetella today consists of nine species, four of which cause human respiratory illness ( B. pertussis, B. parapertussis, B. bronchiseptica, and B. holmesii ). B. pertussis and B. parapertussis are the usual etiologic agents of pertussis, but 86% to 95% of illnesses are caused by B. pertussis. B. pertussis infects exclusively humans. A separate lineage of B. parapertussis has been recovered from sheep as well as humans. In rare instances, B. bronchiseptica , which normally is enzootic in pigs, dogs, cats, rodents, and other animals, has been isolated from humans with pertussis-like cough illnesses. B. holmesii is an occasional cause of pertussis-like illness in adolescents and young adults.
B. pertussis expresses approximately 3121 proteins, many of which are antigenic or biologically active. Since the isolation of B. pertussis in 1906, its microbiologic characteristics have been determined using animal-model and organ-culture systems. The use of these systems has led to many views that are incorrect. Fimbriae (FIM), of which there are two serotypes (types 2 and 3), are protein projections on the surface of B. pertussis that are highly immunogenic. Antibody to FIM causes agglutination of the organism. FIM function as adhesins but may also serve to sustain the attachment established by other adherence factors. Data from two trials in which serologic correlates of immunity were studied indicated that antibody to FIM is important for protection.
Filamentous hemagglutinin (FHA) is a component of the cell wall of all Bordetella spp. It is highly immunogenic and is the dominant attachment factor for Bordetella in animal-model systems. FHA is a component of most diphtheria, tetanus, and acellular component pertussis combination (DTaP and Tdap) vaccines. However, the importance of antibody to FHA and protection from disease is not clear. In the two serologic correlates of immunity studies, FHA did not contribute to protection.
Pertactin (PRN) is an outer membrane protein of B. pertussis that allows it to resist neutrophil-mediated clearance. Vaccine efficacy trials conducted in the 1990s revealed that DTaP vaccines containing PRN in addition to pertussis toxin (PT) and FHA were more effective. Another study revealed that anti-PRN antibodies were required for efficient phagocytosis of B. pertussis by host immune cells. In addition to PRN, there are a number of other similar proteins (autotransporters) that interfere with the host innate immune system.
PT is a critical factor related to mortality in young infants. PT is an adenosine diphosphate (ADP)-ribosylating toxin synthesized and secreted exclusively by B. pertussis. It is an A-B toxin with an enzymatically active A subunit (S1) and a B oligomer (S2-S5) binding portion. PT inactivates G proteins, a process resulting in disruption of signaling pathways and leading to histamine sensitization, enhancement of insulin secretion in response to regulatory signals, and both suppressive and stimulatory immunologic effects in animal model systems. PT also is responsible for leukocytosis with lymphocytosis in B. pertussis infections. In mouse and rat models, PT inhibits chemotaxis and migration of neutrophils, monocytes/macrophages, and lymphocytes to infection sites, and PT also functions as an adhesin in the adherence of B. pertussis to human macrophages and ciliated respiratory epithelial cells. PT is not expressed by B. parapertussis . PT contributes to morbidity in B. pertussis infections, as indicated by the severity of illness, which tends to be greater than that caused by B. parapertussis infection. The A promoter of PT catalyzes ADP-ribosylation of the α subunit of trimeric G proteins, which disturbs metabolic functions of many host cells and leads to a variety of biologic activities, including the lymphocytosis-promoting effect.
Adenylate cyclase toxin (ACT) is an extracytoplasmic enzyme that impairs host immune cell function. Dermonecrotic toxin (DNT) is a cytoplasmic protein that causes skin necrosis in laboratory animals, and tracheal cytotoxin (TCT) causes local damage to respiratory epithelium in hamster tracheal organ cultures and in cultured hamster tracheal epithelial cells.
The type III secretion system (TTSS) allows Bordetella to translocate effector proteins directly into the plasma membrane or cytoplasm of host cells.
The lipopolysaccharide (LPS) of B. pertussis is similar to the endotoxins of other gram-negative bacteria. Its function in disease is unknown, but it may act as an adhesin. LPS is a major cause of reactions to whole-cell pertussis vaccines. LPS is a significant agglutinogen. Antibody to LPS reduces colonization of B. pertussis in the lungs and trachea of mice after aerosol challenge.
In summary, human infections with B. pertussis are due to a number of proteins that are adhesins (mainly FIM but also the B oligomer of PT and LPS) or that interfere with innate immunity (PRN and other autotransporters, ACT and PT). Clinical disease is due to the A subunit of PT and an as yet unidentified toxin or process that causes the cough. B. pertussis infection and clinical illness in humans is noninflammatory in nature, unless there is a concomitant viral or secondary bacterial infection.
Epidemiology and Transmission
Pertussis caused by B. pertussis is an extremely infectious disease, with an estimated 12 to 17 secondary cases produced by a typical primary case in an entirely susceptible population. Attack rates in susceptible household contacts range from 70% to 100%. In the prevaccine era in the United States, the average attack rate of reported pertussis was 157 per 100,000 population. Previous studies, however, suggested that reported cases represent only between 15% and 25% of cases that actually occur. With the introduction and widespread use of whole-cell pertussis vaccines, the attack rate of reported pertussis in the United States fell approximately 150-fold from 1943 to 1976. For the 7-year period from 1976 to 1982, the attack rate in the United States remained between 0.5 and 1.0 per 100,000 population. From 1982 to 2012, the attack rate curve shifted modestly upward and reached a rate of 15.2 per 100,000 in 2012. Possible reasons for the resurgence of reported pertussis that have been suggested are (1) increased vaccine failures resulting from genetic changes in B. pertussis, (2) increased vaccine failures related to vaccines of lessened potency (in general, DTaP [diphtheria-tetanus–acellular pertussis; childhood] vaccines are less efficacious than are DTP vaccines), (3) waning immunity, (4) greater awareness of pertussis, and (5) the availability of better laboratory tests (the use of polymerase chain reaction [PCR] for diagnosis). Pertussis epidemics in the prevaccine era occurred at 2- to 5-year intervals, and these cycles have continued in the vaccine era. As noted by Fine and Clarkson, this continuation of the characteristic cycling during the vaccine era as occurred in the prevaccine era indicates that, although immunization has controlled disease, it has not reduced transmission of the organism in the population.
In the prevaccine era, the following percentages of cases by age were noted in Massachusetts: younger than 1 year, 7.5%; 1 to 4 years, 41.1%; 5 to 9 years, 46.0%; 10 to 14 years, 4.1%; and 15 years and older, 0.9%. Associated with the marked reduction in reported cases of pertussis in the United States resulting from widespread pediatric immunization, a major shift occurred in the percentages by age category. During the period from 1978 to 1981, the age distributions were as follows: younger than 1 year, 53.%; 1 to 4 years, 26.5%; 5 to 9 years, 8.2%; 10 to 14 years, 5.4%; and 15 years or older, 6.5%. In contrast, U.S. data for 2010 revealed the following: younger than 1 year, 15%; 1 to 6 years, 22%; 7 to 10 years, 18%; 10 to 19 years, 20%; and 20 years or older, 25%. Today, pertussis in adolescents and adults is an important source of B. pertussis infection in unimmunized or partially immunized children. ∗
∗ References
B. pertussis infections are endemic in adults, are often not diagnosed as pertussis, and are responsible for cyclic outbreaks in susceptible children.
Transmission is thought to occur by respiratory droplets from a coughing patient that reach the upper respiratory tract of a susceptible person or contaminate environmental surfaces, which then a new host touches and autoinoculates. Transmissibility is greatest early in the illness, that is, during the catarrhal and early paroxysmal phases.
Pathogenesis
After the patient is exposed to B. pertussis , the pathogenesis of infection depends on four important steps: mucosal attachment, evasion of host defenses, local damage, and systemic disease. †
† References .
Although much has been learned about the biologically active and antigenic components of the organism, the exact mechanisms of pathogenesis in human disease are not fully understood. Infection is initiated in the respiratory tract by the attachment of B. pertussis organisms to the cilia of ciliated epithelial cells. Adhesins (PT, FIM, LPS) facilitate this attachment. PRN as well as other autotransporters, often mislabeled as adhesins, have specific biologic activities. Antibody to PRN is associated with lack of attachment. Two household contact studies of serologic correlates of protection both found antibody to PRN to be the most important in preventing pertussis.
Both ACT and PT adversely affect immune cell function and, therefore, allow infection, once initiated, to continue. ‡
‡ References .
PT prevents migration of lymphocytes and macrophages to areas of infection and adversely affects phagocytosis and intracellular killing. ACT enters phagocytic cells and catalyzes excessive production of cyclic adenosine monophosphate, which intoxicates neutrophils and results in a decrease in phagocytosis. Clinical experience and recent studies indicate that PT is the major causative factor in deaths in young infants. PT inhibits G proteins, and this is the cause of the leukocytosis with lymphocytosis. It has been suggested that the extreme leukocytosis results in obstruction in small pulmonary vessels, and this leads to severe pulmonary hypertension. However, it is possible that the inhibition of G proteins in other cells in the lung or heart is the cause of death. In contrast to the situation in B. pertussis infection, leukocytosis with lymphocytosis is not a characteristic of B. parapertussis infection because this organism does not liberate PT.
TCT, DNT, and ACT all have been implicated as contributors to local tissue damage in the respiratory tract. However, there are no data indicating tissue damage in human infections caused by these three toxins.
The cause of the hallmark paroxysmal cough of disease caused by B. pertussis infection is unknown. The paroxysmal cough is unique because of its long duration and memory, returning when people have subsequent respiratory viral infections. PT has been suggested to be the cause of the prolonged cough in pertussis. However, because persistent cough is a major manifestation of B. bronchiseptica infection in dogs and of B. parapertussis infection in children, and because neither organism expresses PT, this hypothesis should be refuted. Recently, Hewitt and Canning have suggested that bradykinin plays a role in the initiation of the cough. During B. pertussis infection, bradykinin production is increased, exhibits a prolonged half-life in the airways, and responsiveness to bradykinin reception activation is increased. Bradykinin activates sensory nerves implicated in cough. However, the Bordetella protein(s) or processes that stimulate the unique coughing in pertussis remain unknown.
Cell-mediated immune function is altered by B. pertussis infection. In some studies, cell-mediated immunity was depressed, whereas in others it was augmented.
Various antibodies develop after exposure of the human host to infection with B. pertussis . The development of agglutinins, hemagglutination-inhibiting antibodies, and bactericidal antibodies has been described. Enzyme-linked immunosorbent (ELISA) techniques have demonstrated class-specific antibodies (immunoglobulin A [IgA], IgE, IgG, and IgM) to many of the specific proteins of B. pertussis. These antibodies develop after infection and, with the exception of IgA antibodies, also after immunization. Neutralizing antibody to PT likewise develops after both infection and immunization. Specific IgA antibodies to PT and FHA can also be demonstrated in nasopharyngeal (NP) secretions and saliva.
At present, both B. pertussis infection and immunization with whole-cell (DTwP) or acellular pertussis (DTaP) vaccines clearly elicit protection of varying degrees and duration against pertussis, neither providing life-long protection. §
§ References .
Studies in young German and American men looking at the rate and mean values of IgA antibodies to pertussis antigens (which result from infection and not from primary vaccination) demonstrated no difference, suggesting that rates of adult infection were the same regardless of personal history of childhood immunization. In Germany, routine childhood immunization was not carried out during the 1970s and 1980s, and pertussis was epidemic.
The nature of immunity in pertussis remains poorly understood. The consensus has been that serum antibodies greater than some unknown concentration to one or more of the pertussis antigens are responsible for protection. However, no serologic correlate of immunity has been accepted. Although several large vaccine trials were carried out in the late 1980s and early 1990s, only two were done in a manner that allowed the study of possible serologic correlates of immunity. In a nested household contact study looking at the roles of IgG antibodies to PT, FHA, PRN, and FIM 2 in children at the time of household exposure to B. pertussis , it was shown that geometric mean antibody values to PT, PRN, and FIM 2 were higher in noncases than in cases; however, in the classification tree and regression analyses, only antibodies against PRN contributed significantly to protection. Similarly, a Swedish study found that higher antibody values to PRN and FIM 2/3 correlated with protection. However, some children get pertussis despite large amounts of antibody to either or both PRN and FIM. Data from a study by Weiss and colleagues suggested that high antibody values to PT can block the protective effect of antibody to PRN and perhaps FIM.
Cell-mediated immune responses to PT, FHA, and PRN also occur. Studies in a murine respiratory infection model suggest that cellular immunity plays an important role in bacterial clearance and augments the effects of antibody by predominantly T-helper 1 (TH1) cell stimulation. Human studies demonstrate a cellular immune response shortly after natural infection with B. pertussis, with PT, FHA, and PRN preferentially inducing the synthesis of TH1 cells. Immunization with a whole-cell pertussis vaccine resulted in a TH1 response, whereas the response to acellular vaccines is more heterogeneous and involves both TH1 and TH2 cells. Persistent memory T and B cells and anamnestic antibody responses are important in long-term immunity.
Immunity developed after having B. pertussis infection or receiving vaccination with a whole-cell pertussis vaccine does not protect against illness caused by B. parapertussis, and, similarly, infection with B. parapertussis does not induce protection against disease caused by B. pertussis. However, in a vaccine efficacy trial in Germany, the results showed some evidence that the acellular pertussis multicomponent vaccine, which contained a large amount of FHA, offered some protection against B. parapertussis infections, whereas the whole-cell vaccine, which contained minimal amounts of FHA, did not.
Pathology
Data on pathology of B. pertussis infections have been determined mainly by postmortem study in fatal cases. ∥
? References .
In a study of autopsy material from 15 infants (≤4 months of age), the consistent histopathologic features were necrotizing bronchiolitis and pneumonia, intraalveolar hemorrhage and fibrinous edema, and abundant intraalveolar macrophages. Angiolymphatic aggregates of mixed leukocytes in the intralobular septa and pleurae were seen in 86% of the specimens. Intact Bordetella organisms were noted in cilia of the trachea, bronchi, and bronchioles and within airways and alveoli. The pathogen was also noted intracellularly in alveolar macrophages and respiratory epithelium. Despite the fact that all patients had severe pneumonia, the ciliated respiratory epithelial cells appeared relatively normal. Six of the patients had evidence of coinfections with one or more other agents (cytomegalovirus, 1 child; respiratory syncytial virus [RSV], 2 children; Streptococcus pneumoniae, 2 children; Streptococcus pyogenes, 2 patients; Moraxella catarrhalis, 1 child; and viridans streptococci, 1 child).
Pathologic changes in the brain and liver have also been described. Microscopic or gross cerebral hemorrhage may be noted, and cortical atrophy has been observed. These changes most likely are the result of anoxic brain damage. In some studies of pertussis encephalopathy, findings suggested meningoencephalitis with perivascular cuffs of lymphocytes within cerebral gray matter and pleocytosis. However, the studies in which inflammation was demonstrated were performed before modern virologic techniques became available. Therefore the neurologic findings in these instances may have been caused by interactions with neurotropic viruses or other infectious agents and were not the result of B. pertussis infection. Fatty infiltration of the liver has been noted in patients with pertussis encephalopathy.
Although B. pertussis infection is a localized infection involving ciliated cells of the respiratory tract, bacteremia has been noted in immunocompromised adults.
Clinical; Bordetella pertussis Infections
Adults, Including Pregnant Women and Mothers
The clinical presentation in adults, including pregnant women, can be nonspecific, including findings of coryza and cough without the characteristic whoop. Several studies have found that from 12% to 30% or more of persons with acute illness with cough of at least 1 to 2 weeks of duration have evidence of B. pertussis infection, whereas it varies from 21% to 86% in those with classic pertussis symptoms, such as paroxysmal cough, whoop, and posttussive vomiting. ¶
¶ References .
Three important points relating to pertussis in adults (including pregnant women) and older children are the occurrence of coryza that does not become purulent, the lack of fever, and the lack of leukocytosis with lymphocytosis that is seen in primary infections in young children. Also important in pertussis in adults is the occurrence of sweating episodes between paroxysms of coughing. Whereas complications of pertussis in adolescents and adults are less common than in young children, hospitalization, pneumonia, and seizures do occur. In addition, urinary incontinence, fractured ribs, herniated intervertebral disc, hearing loss, angina attack, carotid artery dissection and encephalopathy may occur.
In 2008, the Advisory Committee on Immunization Practices (ACIP) presented their recommendations for the prevention of pertussis among pregnant and postpartum women and their infants. In this report, they reviewed the literature relating to pertussis in pregnancy. They concluded that the morbidity of pertussis was not increased among pregnant women compared with nonpregnant women. Granström and associates reported 32 women who had pertussis in late pregnancy. These illnesses were not associated with obstetric complications, and all pregnancies went to term. Reports of adverse effects upon the fetus are rare, and no causal relationship with abnormal fetal development or fetal morbidity has been established. The fetus of one woman with pertussis had an extradural hematoma and the fetus of another woman had tracheal obstruction.
McGregor and associates described three cases of maternal-infant pairs in which the mother-to-newborn transmission of pertussis led to severe pertussis in the child. Each of the mothers described had prolonged severe paroxysmal cough associated with posttussive emesis. In the first case, the mother developed a severe nonproductive whooping-like cough 5 days after delivery, with development of paroxysm and posttussive emesis. In the second case, the mother and a sibling had begun coughing 1 day before delivery. In the third case, immediately postpartum, the mother complained of coryza-like symptoms with a gradually increasing cough with paroxysm.
In another instance, a 22 year-old mother was hospitalized 6 days before delivery because of severe paroxysms of coughing with posttussive emesis. The neonate had the onset of cough on the seventh day of life and subsequently died of the illness.
Multiple studies have demonstrated that the source of infection in infants is usually a family member. #
# References .
In a study of 616 infant cases, the source was identified in 43%. A family member was the source 75% of the time, and the mother was the most common source (32%). Of the source persons, 56% were adults, and 20% were 10 to 19 years of age. In studies of household contacts, asymptomatic infections in family members are common occurrences. Deen and associates found that 52 (46%) of 114 household contacts who remained well had laboratory evidence of B. pertussis infection. In another study, 21 of 399 healthy infants who were controls in a study of sudden infant death syndrome (SIDS) had PCR-positive NP samples. In a study at one hospital during the California pertussis epidemic of 2009 to 2010, 32 infants younger than 3 months were hospitalized with pertussis, and household coughing contacts were reported for 24 out of the 32 patients with pertussis (75%); the patient’s mother was the contact in 10 patients (42%), and a sibling was the contact in 11 patients (46%).
A study of pregnant women in the Netherlands from January 2004 to January 2006 looked at the seroprevalence of Bordetella pertussis infection during pregnancy. Specific antibodies against pertussis toxin were measured in umbilical cord blood samples, maternal blood samples at 12 weeks of gestation, delivery, and 2 months after delivery. In total, 20 (6.3%) of 315 pregnant women in the study had serologic evidence of B. pertussis infection during or shortly before pregnancy, which was much higher than the incidence of reported cases for that age group. They concluded that their findings emphasized that B. pertussis infection often remains unreported, possibly because of a subclinical course of infection or the failure to recognize or report the disease by a general practitioner.
Neonates and Young Infants
Pertussis in neonates and young infants is a unique experience. Its spectrum of clinical manifestations varies by age, immunization status, and the presence or absence of transplacentally acquired antibody. ∗a
∗a References .
Most deaths resulting from B. pertussis infection occur in neonates and early infancy, and morbidity is most severe in infants. †a
†a References .
From 1997 through 2000, 8276 cases of pertussis were reported in infants in the United States. Of this group, 59% were hospitalized, 11% had pneumonia, 1% had seizures, 0.2% had encephalopathy, and 0.7% died. Eighty-seven percent of these infant cases occurred in children younger than 6 months. Clinical characteristics of pertussis in young infants as described in several recent studies are presented in Table 21-1 . ‡a
‡a References .
TABLE 21-1
Clinical Characteristics of Pertussis in Young Infants
Data from references , and .
| Smith and Vyas, 2000 | Castagnini and Munoz, 2010 | Taffarel et al, 2012 | Murray et al, 2013 | Berger et al, 2013 | Rocha et al, 2013 | |
|---|---|---|---|---|---|---|
| Number of points | 9 | 33 | 41 | 31 | 127 | 18 |
| Age | <7 weeks (all PICU) | ≤30 days | Mean age, 2.38 months | ≤90 days (all PICU) |
105 (83%) were <3 months | ≤2 months |
| Cough | 9 (100%) | 32 (97%) | N/A | N/A | 125 (98%) | 16 (89%) |
| Apnea | 5 (56%) | 19 (58%) | N/A | N/A | 62 (49%) | N/A |
| Cyanotic spell | N/A | 30 (91%) | N/A | N/A | 92 (72%) | 2 (11%) |
| Seizures | 5 (56%) | 1 (3%) | N/A | 4 (13%) | 14 (11%) | 1 (6%) |
| O 2 need | N/A | 27 (82%) | 24 (59%) | N/A | 104 (82%) | 18 (100%) |
| Mechanical ventilation | 9 (100%) | 9 (27%) | 31 (76%) | 8 (26%) | 55 (43%) | 6 (33%) |
| Hyper/hypotension | 7 (78%) | N/A | 27 (66%) | 4 (13%) | 45 (35%) | 4 (22%) |
| Pulmonary hypertension | 1 (11%) | N/A | 16 (39%) | 6 (19%) | 16 (13%) | 4 (22%) |
| Need for dialysis | N/A | N/A | 4 (10%) | 3 (10%) | 20 (16%) | N/A |
| Died | 6 (67%) | 0 (0%) | 17 (41%) | 4 (13%) | 12 (9%) | 3 (17%) |
N/A, Not available; PICU, pediatric intensive care unit.
B. pertussis infection in neonates is particularly severe, with a death rate of between 1% and 3%. §a
§a References .
Clinical characteristics of pertussis in early infancy are summarized in Box 21-1 . A common initial finding is apnea, and typical coughing may not be observed. Seizures in association with apnea caused by hypoxia occur frequently. Severe pulmonary hypertension is a relatively common problem in pertussis in the first 4 months of life. ∥a
?a References .
The severity of disease and the risk of death correlate directly with the white blood cell (WBC) count and, in particular, the number of lymphocytes. ¶a
¶a References .
WBC counts in the range of 30,000 to more than 100,000 cells/mm 3 are common findings. A recent California Department of Public Health study has provided useful information relating to risk factors for pulmonary hypertension and death. This study involved 31 infants younger than or equal to 90 days of age with pertussis admitted to a pediatric intensive care unit (PICU) in Southern California. Infants with pulmonary hypertension or who died had higher WBC counts (>30,000 cells/mm 3 ), more rapid pulse (>170) and respiratory rates (>70), and were more likely to have pneumonia than those who did not have pulmonary hypertension or die. In addition, in those with pulmonary hypertension or who died, the WBC count rose more rapidly, and the pulse and respiratory rates attained higher levels sooner than in those without pulmonary hypertension or death.
Box 21-1
Clinical Features of Pertussis in Young Infants
Data from references , and .
- 1.
Initially, infant looks deceptively well; coryza, sneezing, throat clearing, no fever, mild cough
- 2.
Paroxysmal stage: gagging, gasping, eye bulging, bradycardia (tachycardia occurs in severe illness), cyanosis, vomiting
- 3.
No or minimal fever unless complicated by a secondary bacterial infection or a concomitant viral infection
- 4.
Leukocytosis with lymphocytosis
- 5.
Apneic episodes
- 6.
Seizures
- 7.
Respiratory distress
- 8.
Pneumonia and pulmonary hypertension
- 9.
Shock/hypotension with renal failure
- 10.
Death associated with pulmonary hypertension and pneumonia
- 11.
Adenovirus or respiratory syncytial virus coinfection can confuse picture
Coinfections with respiratory viruses (RSV, adenovirus, influenza viruses) and respiratory bacterial pathogens ( S. pneumoniae , Haemophilus influenzae ) are relatively frequent. #a
#a References .
A recent study by Berger and associates described 127 children hospitalized with critical pertussis and again found worse severity of illness in patients younger than 3 months. Eighty-three percent of patients were younger than 3 months. The median WBC count was significantly higher in those requiring mechanical ventilation (35,200/mm 3 vs. 26,100/mm 3 ), those with pulmonary hypertension (68,400/mm 3 vs. 25,100/mm 3 ), and nonsurvivors (66,300/mm 3 vs. 26,100/ mm 3 ).
B. pertussis infection has been noted in association with SIDS, but whether a cause-and-effect relationship exists is not clear. Nicoll and Gardner, in a study in England, found that many deaths attributed to SIDS were, in fact, related to B. pertussis infection. Using PCR, B. pertussis DNA was noted in NP specimens from 9 (18%) of 51 infants who had sudden, unexpected deaths. In a subsequent study, specimens were collected for PCR from 254 infants who experienced sudden, unexplained deaths and from 441 healthy matched control subjects. The rate of PCR-positive results in the sudden death cases was 5.1%; it was 5.3% in the control group. In a careful follow-up histopathologic study with unique immunohistochemical staining of specimens from a subset of these fatal cases, no evidence of specific B. pertussis pulmonary infection or pathologic features was found.
Aside from prevention of disease, early detection can make a difference in outcomes. However, early and/or presenting symptoms in neonates are not often obvious. A recent study comparing hospitalized patients younger than 3 months with confirmed pertussis versus RSV or influenza indicated several features that should alert clinicians to pertussis, including paroxysmal cough, posttussive emesis, lack of congestion, and lack of fever.
Other Bordetella Infections
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here