Asthma is a chronic airway inflammatory disease characterized by bronchial hyperresponsiveness and reversible airflow obstruction. A cause of health problems globally for all ages, asthma affects approximately 300 million people worldwide. Its incidence rate, treatment costs and burden on health care systems in many developed nations such as China, Australia, New Zealand, Canada and the United Kingdom, are increasing [1]. The Global Initiative on the Prevention and Treatment of Asthma (GINA) defines cough variant asthma (CVA) as a special type of asthma without wheezing or shortness of breath, where coughing is the sole or main symptom [1]. CVA in children is a particular type of asthma, predominantly characterized by a persistent cough. Children with CVA have no typical clinical symptoms and are easy to misdiagnose, resulting in their being unable to receive standardized treatment. The disease can then develop into typical asthma, hence affecting the child’s growth. CVA in children can be prevented, but it is difficult to cure, and always causes repeated attacks. Therefore, discovering the hazard factors of CVA is of great importance in the prevention and provision of relevant treatment for it.

In contrast, the total number of children diagnosed with congenital heart disease (CHD) has been rising due to improved diagnosis. However, there has been no improvement in its treatment. Over the past 60 years, great progress has been made in the medical and surgical care of these children. However, specialists are facing challenges as these children develop new complications [2]. They often need long-term follow-up for lifelong heart problems and decades of reoperations to carry out surgical repairs that may even lead to various respiratory diseases. For children with CHD complicated by CVA, it is necessary to carry out further studies into the likelihood of incidence and recurrence rates increasing. At the same time, the relationship between the severity of airflow obstruction and cardiovascular disease’s morbidity and mortality has been well established. However, the relationship between CVA characterized by variable airway obstruction and cardiovascular disease is not clear, and data reported on the relationship between asthma and cardiovascular disease have not been confirmed [3].

In addition, considering the gradual increase in the incidence and recurrence rates of CVA, children with CHD are of great interest. The effect of CHD on the recurrence rate of CVA in children has rarely been studied till now.

A total of 489 children aged between one and 14 years with CVA were recruited from the hospital-based outpatient Respiratory Disease Clinic during the period from September 2018 to March 2019. The diagnostic criteria of CVA were based on the Guidelines for the Diagnosis and Prevention of Bronchial Asthma in Children, 2016 [4]. Figure 1 depicts the cohort selection process. We identified 67 cases of CHD complicated by CVA from their medical history and color doppler ultrasound examination from these children. The date of the CVA diagnosis was set as the index date. Index date is the time when the children were enrolled, and the starting time of each group of experimental studies. The starting time of each group of patients may be different. We then frequency matched the 134 children with CVA but without CHD, in a ratio of 1:2 according to sex, age and index date. The Medical Ethics Committee of Guangdong Provincial People's Hospital reviewed the project and obtained the written informed consent of the parents or guardians. All the children were given routine blood tests, Mycoplasma pneumoniae antibodies and allergen specific IgE tests, and their family history was collated. After their clinical symptoms had been controlled, they were followed up for 12 months. During the follow-up period, drug treatments and allergen avoidance were standardized. The recurrence of CVA was assessed using the cough visual analogue scale(VAS) score, the total symptom score, the daytime and nighttime symptom score, the children asthma control test(C-TAC) score, and the test for respiratory and asthma control in kids(TRACK) score.

We screened 67 CVA children with CHD and 134 children with non-CHD from 2018 to 2019. During the follow-up period, there were four occurrences of censored data in the CHD cohort and seven in the non-CHD group. Table 1 shows a comparison of demographic characteristics and clinical manifestations between the two cohorts. The gender and age distributions of the cohorts were similar. Of the children, 55.2% were male, 40.3% were aged 1–3 years, 28.4% were 4–6 years old and 31.3% were 7–14 years of age. There were no significant differences in family history, laboratory examinations or treatments between the two cohorts at the baseline.

The cumulative hazard of recurrent CVA was significantly higher in the CHD cohort (P = 0.001, log-rank test) than in the non-CHD group (see Fig. 2).

Table 2 shows univariate and multivariate analysis of CVA predictors. From the univariate analysis of CVA predictors, CHD classification, MP infection and IgE sensitization were significant differences, while most predictors did not differ between the two virus cohorts. Overall, the multivariable Cox proportional hazards model demonstrated that the recurrence hazard of CVA in the CCHD group was significantly higher than that in the non-CHD cohort (aHR = 3.281; 95% CI 1.648–6.530; P = 0.001) after adjusting for CHD classification, MP infection and IgE sensitization; while the recurrence hazard of CVA in the SCHD group was also significantly higher than that in the non-CHD cohort (aHR = 2.555; 95% CI 1.739–3.752; P = 0.000). Further, the recurrence hazard of CVA with IgE sensitization (aHR = 2.172; 95% CI 1.482–3.184; P = 0.000) had a higher hazard than those without IgE sensitization; the recurrence hazard of CVA with MP infection (aHR = 1.777; 95% CI 1.188–2.657; P = 0.005) had a higher hazard than those without the MP infection.

Figure 3 shows the hazard ratios of non-CHD versus CHD cohorts for risks of all-cause mortality in the prespecified subgroups. The hazard of recurrent CVA was higher in male (HR = 2.69; 95% CI 1.67–4.32; P < 0.01) than in female children (HR = 2.12; 95% CI 1.23–3.65; P < 0.01). Using age stratification, CVA in CHD patients had the highest recurrent hazard ratio among those aged 1–3 years (HR = 2.56; 95% CI 1.45–4.53; P < 0.01). Children with CHD who had an elevated eosinophil ratio (HR = 2.51; 95% CI 1.30–4.82, P < 0.01) had a higher hazard of recurrent CVA than those with a normal eosinophil ratio (HR = 2.40; 95% CI 1.57–3.68; P < 0.01), but there was no significant interaction between the hazard ratios of non-CHD versus CHD cohorts for risks of all-cause mortality in relation to the hazard of recurrent CVA (P for interaction > 0.05,see Fig. 3).

On the one hand, among the four subgroups stratified by MP infection, the children with CHD and MP infection had the highest recurrence rate (P = 0.001, log-rank test, see Fig. 4). But there was no significant interaction between CHD and MP infection in relation to the hazard of recurrent CVA (P for interaction = 0.512, see Fig. 3). On the other hand, among the four subgroups composed according to IgE sensitization, the hazard of recurrent CVA in children with CHD along with IgE sensitization was highest (P = 0.000, log-rank test, see Fig. 5), but there was no significant interaction between CHD and IgE sensitization in relation to the hazard of recurrent CVA (P for interaction = 0.414, see Fig. 3).

Table 3 shows that the aHR of these CVA children decreased from 1.852 (95% CI 1.246–2.751; P = 0.001) within six months to 0.691 (95% CI 0.088–5.404; P = 0.725).

CVA is a particular form of asthma, having a close pathogenesis to typical asthma. Asthma is a disease characterized by infiltration of the mucositis cells in the airway and thickening of the subepithelial basement membrane in the airway [5]. First, we found that those CVA patients with CHD had a higher hazard of recurrence, and that the recurrence hazard for children with CCHD was even higher. We also found that IgE sensitization and MP infection could increase the hazard of CVA recurrence. The effect of CHD on CVA recurrence has hardly been studied previously.

It has been reported that the mechanism for CVA caused by heart disease is bronchial vascular congestion, resulting in bronchial oedema and thickening [6]. In addition, it has been reported that left ventricular dysfunction can lead to abnormal pulmonary function, such as airway hyperresponsiveness or restrictive and obstructive dysfunction [7]. Highly reactive heart failure in patients is thought to be caused by bronchiectasis and submucosal oedema. The incidence of diastolic dysfunction associated with obstruction was over four times higher than that in the control group [8]. This is caused by oxidative stress factors. Although according to current understanding, volume overload in heart disease may also be a cause of inflammation. This additionally leads to airway remodeling [9]. In addition, children with CHD often have bronchial hyperresponsiveness, and this, especially when there is left-sided heart failure, can also damage the lungs. It can lead to pulmonary vascular remodeling, increased pulmonary vascular resistance, and alveolar wall thickening. To some extent, these changes have a protective effect and can prevent pulmonary oedema. Over a long period, however, they may lead to pulmonary hypertension and affect lung function [10].

It is generally believed that CHD is associated with recurrent asthma attacks in children. For example, there is a high prevalence of airway hyperresponsiveness in patients with atrial septal defects (ASD), and it is suggested that airway hyperresponsiveness may be a possible mechanism for recurrent attacks of CVA. Unlike bronchial asthma, 11 asthma-like symptoms associated with ASD are not necessarily experienced every few hours or days, but can be progressive [11]. This better understanding of ASD-associated dyspnea could potentially prevent delays to the treatment of unrepaired ASD. It may also prevent a hazard of complications associated with long-standing right ventricular volume overload and delayed diagnosis of pulmonary function damage after long-term operation. In clinical follow-ups of adult patients with ASD, symptoms and lung function should be monitored for a long time before and after any operations [12]. Therefore, the effect of CHD plays an important role in the recurrence of CVA in children.

In addition, according to the complexity of cardiac malformation, CHD can be divided into SCHD and CCHD. CCHD was found to cause a higher hazard of recurrent CVA than SCHD. Complex blood flow relationships may lead to pulmonary vascular remodeling, an increase in pulmonary vascular resistance and alveolar wall thickening, and pulmonary tracheal remodeling, thus increasing the hazard of CVA recurrence. It has been reported that in some heart defects, especially in the left to right shunt, reversible airway obstruction may be caused by abnormal or dilated vascular anatomical compression due to missing pulmonary valves, vascular rings, or pulmonary slings [13].

The second important finding of this study was that children with IgE sensitization had an increased recurrence rate of CVA. Ciprandi et al. found that allergic children who identified with a positive skin prick test or history of allergic diseases had more frequent and severe respiratory tract infections than non-allergic children [14]. The underlying mechanism for the positive association between IgE sensitivity and pneumonia is still unclear. It may involve a reduction in the Type 1 helper cell (T_h1) response in IgE sensitized patients, which may be important for the prevention of infections and mucositis, and may also involve a reduced defense against infectious factors associated with general allergic diseases, such as atopic dermatitis [15]. This suggests that IgE sensitization may not have a causal relationship with susceptibility to infection, but that IgE sensitization and infection share a common pathway, or that people who are more susceptible to infection may develop atopic diseases more frequently while maintaining a tendency to infection.

In this study, children with MP infection had a higher hazard of recurrent CVA. Relevant research results abroad have shown that MP infection can induce CVA, and MP is considered one of the pathogenic factors for the acute onset, difficult remission, and deterioration of CVA. However, there is no unified theory on the induction mechanism. Some scholars believe that because MP infection is a specific antigen, it can lead to IgE-mediated airway inflammation and airway hyperreactivity through delayed allergic reaction and immediate allergic reaction or high interference and disorder on T-lymphocyte subsets, and finally lead to CVA [16]. Infants with MP infection, particularly with IgE sensitization during infancy, had the highest hazard [17]. Therefore, MP infection may aggravate the recurrence of CVA.

During the one-year follow-up period, the hazard of recurrent CVA in children with CHD was highest in the first six months, while the aHR decreased after six months. This indicated that time-dependent influencing factors were excluded in this study, and that CHD was a high-hazard factor for CVA recurrence within the first six months.

In a prospective cohort study of 201 children with CVA, we found a significantly increased hazard of recurrent CVA in children with CHD, especially in CCHD children. In addition, the hazard of recurrent CVA can increase for children with IgE sensitization or MP infection. Clinicians at outpatient clinics need to be aware of this association and thus pay more attention to the hazard of recurrent CVA in children with CHD.