Analysis of clinical characteristics of pertussis in 184 infants and children: a retrospective study

Backgroud: The incidence of pertussis shows an increasing trend in recent years, but some clinicians often lack sucient understanding of the clinical characteristics and risk factors for severe pertussis, and more effective measures should be taken to reduce the incidence and mortality of pertussis in young infants. Methods: A retrospective study was conducted, and 184 infants and children with pertussis who had been hospitalized in the Department of Pediatrics of Beijing Ditan Hospital aliated with Capital Medical University from January 2016 to December 2017 were included in the study. Clinical data of the patients were collected and the clinical characteristics were statistically analyzed. Results: Among the 184 patients, 41.85% were infants < 3 months of age, and 65.22% were not vaccinated against pertussis. There were 22 critically ill children, among whom 4 died, and compared with moderate cases, they had a higher proportion of children younger than 3 months of age and infants not vaccinated against pertussis (63.64% vs. 38.89% and 100% vs. 60.49%, respectively); a higher proportion of children with pulmonary consolidation (77.27% vs. 1.85%); higher leukocyte count(× 10 9 /L , 35.80 ± 20.53 vs 19.41 ± 8.59); and a higher proportion of children with severe hyperleukocytosis (18.18% vs. 0%, respectively) (P<0.05). Conclusions: 1. Infants aged < 3 months who were not vaccinated for pertussis tended to be infected with pertussis and had severe pertussis. Adjusting the immunization strategy to vaccinate pregnant women to produce protective effects for the infants should be considered. 2. Severe pneumonia and hyperleukocytosis are the main mechanisms underlying severe pertussis and the key points of prevention and diagnosis of severe pertussis.

produce protective effects for the infants should be considered. 2. Severe pneumonia and hyperleukocytosis are the main mechanisms underlying severe pertussis and the key points of prevention and diagnosis of severe pertussis.

Background
With the development of modern medicine, the incidence of many infectious diseases has signi cantly decreased, along with their threat to human health. However, the incidence of pertussis (whooping cough) shows an increasing trend. This phenomenon is known as "pertussis resurgence." Severe pertussis can cause sudden infant death. This is almost completely unavoidable, even in developed countries such as the United States and France [1,2]. Currently, the reported mortality rate of pertussis is 1.2-3.0% [3]. There are two unique circumstances contributing to this. First, this mortality occurs in part because the pathogenesis and lethality mechanisms of pertussis are not completely clear. Second, some clinicians often lack su cient understanding of the clinical characteristics and risk factors for severe pertussis and thus do not focus their attention on these factors.
In the last few years, the number of pediatric patients with pertussis treated by our hospital has increased. This has included many severe cases and deaths. In this study, we employed a retrospective study design to analyze the clinical data of pediatric patients with pertussis treated by the department of pediatrics at Ditan Hospital in the 2 years of 2016 and 2017 in order to examine the clinical characteristics of pertussis and provide a clinical basis for the prevention and treatment of severe pertussis.
Materials And Methods

Study subjects
This was a retrospective study which had been approved by the ethics committee of Beijing Ditan Hospital, Capital Medical University (approval number2018-6-12B). 184 pediatric patients with pertussis, aged 0-6 years, who were treated at the Department of Pediatrics at Ditan Hospital from January 2016 to December 2017, were consecutively concluded as study subjects. All these infants met the diagnostic criteria for pertussis (see below).

Methods
2.1. The diagnostic criteria for pertussis were as follows [4]: 1) 0-3 months of age: no fever or low fever, cough with increasing frequency and severity, plus one of the following: cough with the characteristic whoop, apnea, vomiting after cough, cyanosis, convulsions, pneumonia, and close contact with patient with long-term fever-free cough (usually a family member); or only paroxysmal apnea, cyanosis, and convulsions without a cough. 2) 4 months to 9 years old: no fever or low fever, paroxysmal cough ≥ 7 days, non-purulent rhinitis, and one of the following: cough with the characteristic whoop, vomiting after coughing, apnea, convulsions, pneumonia, exacerbated symptoms at night, and close contact with patient with a long-term fever-free cough.
Laboratory diagnostic criteria: Bordetella pertussis was cultured in the laboratory, Bordetella pertussis was detected by polymerase chain reaction (PCR), or paired serological test was positive (immunoglobulin G (IgG) titer of single enzyme linked immunosorbent assay (ELISA) test was signi cantly increased > 80-100 U/ml).
In this study, the pathogenic diagnosis was performed using a Bordetella pertussis nucleic acid detection kit, and deep sputum samples of the infants were collected for testing.
Infants with pertussis who also had hypoxemia, recurrent apnea, cardiovascular dysfunction, or pertussis encephalopathy were considered to have severe pertussis.
2.2. Treatment: Pertussis was treated with intravenous instillation of azithromycin at a dose of 10 mg/kg body weight, once a day for 3 days. After a 4-day interval, a second treatment course was performed if needed with the same dosage and treatment course. Azithromycin was administrated orally to neonates less than 28 days old, with the same dosage and course of treatment as above. According to the clinical manifestations, chest imaging ndings, and sputum bacterial culture results of the infants, if the infants were found to have other bacterial infections, other antibiotics were added as needed. If the infants had respiratory failure, myocardial failure, pertussis encephalopathy, or other organ damage, appropriate symptomatic supportive treatments were administrated.
2.3. Clinical data collection and specimen collection: The patients' age, sex, and vaccination status were collected, and blood and sputum samples were collected under sterile condition. The blood routine, alanine aminotransferase (ALT) and creatine kinase isozyme (CK-MB) were tested. When necessary, arterial blood gas analysis was performed. Bordetella pertussis was tested in sputum specimens using a Bordetella pertussis nucleic acid detection kit. Chest computed tomography (CT) and X-rays were used to evaluate the pulmonary infection status of the infants, and the complications (respiratory failure, heart failure, apnea, pertussis encephalopathy, pneumonia, liver damage, and myocardial damage) were evaluated using clinical manifestations and laboratory tests. Antibiotic use, length of hospital stay, and clinical data of blood routine, C-reactive protein, procalcitonin, and sputum bacterial culture results were collected.
3. Statistical Methods SPSS 22.0 software was used for data analysis. Quantitative data such as age, leukocyte count, platelet count, and length of hospitalization were expressed as ± s. For comparison between moderate and severe cases, the t-test was used for normally distributed data, while the rank sum test was used for non-normally distributed data. Qualitative data such as pulmonary consolidation, fever, non-vaccination with pertussis vaccine, and C-reactive protein elevation were expressed as rates. The chi-square test was used for comparison between the two groups.

General information
From January 2016 to December 2017, we enrolled 184 pertussis patients at our hospital, of which 102 were males and 82 were females. The youngest was 23 days old and the oldest was 4 years old. The mean age was 4.50 ± 3.95 months. 77 (41.85%) were < 3 months of age; 120 (65.22%) were not vaccinated against pertussis. 162 were moderate cases, and 22 (11.96%) were severe cases who were complicated with at least one of the following: respiratory failure, heart failure, recurrent apnea, or pertussis encephalopathy Table 1 .  Cases of death(n) 4(2.17%) * The normal range of neutrophilia percentage between the age of 6 days to 4 years is 30-35%, that of C-reactive protein is <5mg/L, and that of procalcitonin is<0.25ng/L. The columns labeled "n" in the brackets shows numbers of children with pertussis with various complications, with abnormal results of laboratory tests, special treatments, or treatment outcomes.

Clinical characteristics of severe pertussis
There were 22 severe cases, and 18 of them had respiratory failure, 16 had heart failure, 11 had recurrent apnea, and 3 had pertussis encephalopathy (table 2). Compared with the children with moderate cases, the 22 severe cases were younger, 63.64% were < 3 months of age, and the proportion of infants < 3 months of age was signi cantly higher than that of patients with moderate cases. None of the severe cases had been vaccinated against pertussis, showing a signi cantly higher rate of non-vaccination than in the children with moderate cases. The incidence of pulmonary consolidation and fever, and the percentage of children with elevated neutrophil percentage, elevated C-reactive protein and procalcitonin levels were higher, and the positive rate of sputum bacterial culture was signi cantly higher than the moderate cases. Moreover, the white blood cell count of severe cases was higher than those of moderate cases, and the proportions of patients with severe hyperleukocytosis (white blood cells > 50*10E9/L) were also signi cantly higher in the severe cases than in moderate cases. The length of hospitalization was signi cantly longer, and the proportions of using 3 or more antibiotics and special-grade antibiotics were also higher in the severe cases than in moderate cases (P<0.05, Table 3).

Etiological analysis of pertussis-complicated pneumonia
Of the 184 children with pertussis, 23 had positive sputum bacterial culture (12.50%), 15 in the moderate group and 8 in the severe group. The pathogens of the moderate group were common community infectious pathogens, such as Staphylococcus aureus, Escherichia coli, Haemophilus in uenzae, and Streptococcus pneumonia, while Gram-negative bacilli infections were more common in the severe group. These bacilli included Acinetobacter baumannii, Pseudomonas aeruginosa, Enterobacter cloacae, and Klebsiella pneumoniae (Table 4). Among them, the sputum cultures of 4 severe cases were found to contain multi-drug resistant bacteria (2 cases of Acinetobacter baumannii, 1 case of Pseudomonas aeruginosa, and 1 case of Stenotrophomonas maltophilia), and 2 and more types of bacteria were cultured in the specimens of 3 severe cases

Clinical characteristics of 4 deaths
Of the 22 children with severe pertussis, 4 died, all of them were infants younger than 6 months of age and 2 younger than 3 months of age. None of them had been vaccinated with pertussis vaccine. The white blood cell count of 3 cases was > 50 × 10 9 /L, with the highest count of 106.68 × 10 9 /L. All 4 cases were complicated with severe pneumonia (pulmonary consolidation), the percentage of neutrophils was high (44.00-81.54%), C-reactive protein was higher than normal in 3 cases (21.1-235.5 mg/L, with normal value of < 5 mg/L), and procalcitonin was high in 2 cases (1.05-8.22 ng/ml, with normal value of < 0.25 ng/ml). All had fever during the course of the disease. One case had positive sputum bacterial culture, which was Enterobacter aerogenes. The drug sensitivity test of the Bordetella pertussis isolated from a 56-day-old infant showed resistance to macrolides. The causes of death of 2 were respiratory and heart failure, 1 was septic shock, and 1 was acute respiratory distress syndrome (ARDS). (Table 5) Causes of death septic shock respiratory and heart failure ARDS respiratory and heart failure # The drug sensitivity test of the Bordetella pertussis isolated from this patient showed resistance to macrolides.
This table shows the general information and clinical data of 4 deaths in severe pertussis cases.

Discussion
Pertussis can occur in all age groups, with the highest incidence in infants and young children, and infants under 1 year of age have relatively severe conditions. The infants included in this study were hospitalized pertussis patients, with an average age of 4.50 ± 3.95 months. Most of them were young infants with relatively severe conditions. In this study, 65.22% of the infants were not vaccinated against pertussis, and 41.85% were infants younger than 3 months. None of the severe infants were vaccinated, of whom 63.64% were younger than 3 months of age. The study shows that nearly half of the patients experienced onset of pertussis before the standard age of pertussis vaccination in China (3 months of age), and the proportion of unvaccinated infants was signi cantly higher in the severe cases than in the moderate cases, which was consistent with the results of previous studies [5]. In order to decrease the rate of pertussis overall and reduce the number of severe cases, the Chinese government may should consider to adjust the standard immunization schedule. Because IgG antibodies can cross the placenta, almost all IgG antibodies present in infants 6 months old and younger arise from the mother. A set of studies showed that when acellular pertussis vaccine was given to pregnant women in the second or third trimester, especially at 27-36 weeks of pregnancy, enough antibodies against Bacillus pertussis, such as the anti-pertussis toxin and anti-lament hemagglutinin antibodies, were detected in the blood of the neonates, and the antibodies persisted and gradually declined until completely attenuated at the age of 3 months [6,7,8]. Therefore, acellular pertussis vaccine for pregnant women could protect infants from pertussis infections before the standard age of pertussis vaccination (the age of 3 months). So far, acellular pertussis vaccination in pregnancy has been implemented in Brazil, Argentina, Britain, America, Belgium and New Zealand. An observational study showed that the strategy produced up to 90% protective effect for the infants due to its two effects, one was that the speci c antibodies arising from the mother, and the other was protecting the mother from infections that could reduce maternal-neonatal transmission [9,10,11,12]. In Argentina the deaths of pertussis cases deceased 87% in 2013 compared with that in 2011, since the country implemented the strategy of acellular pertussis vaccination in pregnancy in 2012 [ 13].
Pneumonia is the most common complication of pertussis. In this study, more than 80% of hospitalized infants with pertussis also had pneumonia. Some infants had fever, with increased neutrophils, C-reactive protein, and procalcitonin levels, indicating the presence of other bacterial infections. Moreover, 10% of the infants had pulmonary consolidation, suggesting severe infection. All severe cases also had pneumonia, and nearly 80% had pulmonary consolidation, more than 80% had respiratory failure, and more than 60% required ventilator-assisted breathing. All fatal cases were complicated by severe pneumonia. In this way, pneumonia is not only the most common complication of pertussis but also the leading cause of severe cases and death of pertussis [ 14,15,16].
Increased white blood cell count is a characteristic manifestation of pertussis. The white blood cell count of children in the severe group was signi cantly higher than that of the moderate group, and the white blood cell count was particularly high in fatal cases. Due the poor ability of white blood cells to change shape, they tend to obstruct stenosed alveolar capillary beds. White blood cells usually take 10-15 times longer than erythrocytes to pass through these vessels, resulting in embolism due to leukocyte clumps, causing hypoxemia and pulmonary hypertension [17,18]. This affects cardiac function and causes heart failure in severe cases. Severe hyperleukocytosis (> 50 × 10 9 leukocytes/L) is an independent risk factor for malignant pertussis (life-threatening severe pertussis) [19,20]. Many studies have indicated that when routine treatment was given even though the peripheral leukocyte count was > 100 × 10 9 /L in pertussis patients, and no measures to lower leukocyte counts were carried out, death resulted in all cases [21,22]. In this study, there were 4 infants the in the severe group whose white blood cell counts were > 50×10E9/L, of whom 3 died. This awaits con rmation in future work and exploration of related targeted treatment. Studies have shown that plasma exchange and leukapheresis can reduce leukocyte count and signi cantly improve fatal hypoxemia and pulmonary hypertension [23,24]. Some researchers have used carbon monoxide and sildena l to dilate pulmonary blood vessels and reduce pulmonary hypertension; some e cacy was observed [25].
Pertussis not only seriously endangers the life and health of infants, and also creates a heavy burden on families and society. Severe pneumonia and hyperleukocytosis are key issues in the prevention and treatment of severe pertussis. Adjusting the immunization strategy to vaccinate pregnant women has a clear effect on reducing the incidence and mortality of pertussis in young infants, which should be considered by the government.

Conclusions
1. Infants aged < 3 months who were not vaccinated for pertussis tended to be infected with pertussis and had severe pertussis. Adjusting the immunization strategy to vaccinate pregnant women to produce protective effects for the infants should be considered. This retrospective study has been approved by the ethics committee of Beijing Ditan Hospital, Capital Medical University (approval number2018-6-12B). Verbal informed consent was obtained from the legal guardians of the patients in this study because this is a retrospective study and the patients had been discharged when the study was carried out. This procedure was approved by the ethics committee of Beijing Ditan Hospital, Capital Medical University.

Consent for publication
Consent for publication have been obtained from the legal guardians of the patients included in the study.
Availability of data and materials