Introduction

Chronic granulomatous disease (CGD) is a rare inborn error of innate immunity in which a defect in one of the subunits of NADPH oxidase results in an impaired respiratory burst of phagocytic cells, causing recurrent life-threatening infections, inflammatory complications, and autoimmunity [1,2,3,4,5]. The implementation of routine antimicrobial prophylaxis and the advent of azole antifungals has significantly improved overall survival [6, 7] and reduced the estimated annual infection rate around to 0.3 per year [5, 8]. Hematopoietic stem cell transplantation (HSCT) is at now the only curative treatment for CGD as has achieved in last decade good outcomes with both matched siblings and unrelated donors. However, there is still debate about optimal timing and condition regimen of HSCT [9,10,11,12,13].

Despite improvement in management of CGD, the median age of death remains around III–IV decades of life, and patients still have to cope up with a range of disease-related symptoms for their entire life [3, 5, 8, 14]. Symptoms may be related to the infection itself, often resulting in hospitalization for prolonged courses of treatment, relapse of inflammatory comorbidities, and iatrogenic effects from medications. For this, patients become increasingly debilitated with advancing age, with a negative impact on their health-related quality of life (HRQoL). Moreover, many of the CGD patients may face with cognitive and social problems in peer relationships or school/work activities [15]. Quantifying the HRQoL in primary immunodeficiency (PID) conditions has recently began as an effort to document the patients’ perceived burden of disease, tracking changes in health over time, and assessing the effects of therapeutic intervention [16, 17]. Moreover, HRQoL has been shown to be a reliable predictor of mortalities in primary immunodeficiency [18]. For CGD, few data are available on HRQoL and they were collected only by generic instruments. Recently published data from the UK cohort has documented that children with CGD have poor HRQoL compared with healthy peers [19], whereas, to the best of our knowledge, no data has been provided on adults with CGD.

The aim of our study was to evaluate HRQoL and psychological health in Italian CGD children and adults undergoing both conventional and HSCT. HRQoL status was analyzed with regard to clinical phenotype, familiar background, and treatment. CGD HRQoL status was also compared with normative data and to individuals with other chronic diseases. The underlying hypothesis was that CGD patients have a poor HRQoL and that, in addition to clinical assessment, mental and psychological aspects should be considered because of their contribution to the burden of the disease.

Methods

Population

Eligible patients were CGD children and adults regularly followed up in 11 PID centers across Italy with a confirmed diagnosis according to the revised ESID registry criteria (http://www.esid.org/Working-Parties/Registry/Diagnosis-criteria). Participants were identified by the Italian CGD Registry. The registry was established in 2001 by Italian centers belonging to the Italian Association of Pediatric Hematology and Oncology (AIEOP)–Italian network for Primary Immunodeficiencies (IPINET) and it has enrolled 109 patients up to now [5].

Study Design

Participants or their caregivers were recruited during routine clinic appointments. After providing written informed consent, patients or their parents were provided with questionnaires. In CGD who underwent HSCT, questionnaires were administered at least 1-year post-HSCT. Clinical data related to CGD complications and treatments were gathered from medical records. Social status was assessed by individual interview. The clinical status was scored ranging from severe to very good considering the combination of 1) hospitalization in the last 12 months, 2) rate of infections in the last 12 months, and 3) permanent organ damage, as shown in Table 1.

Table 1 Clinical status staging definition by permanent organ damage, hospital admission, and rate of infections in the last 12 months
Full size table

Children were evaluated by the Pediatric Quality of Life Inventory™ version 4.0 (PedsQL) [20] and the Strengths and Difficulties uestionnaire (SDQ) [21]. Adults were evaluated by 12-item Short Form health survey (SF-12) [22]. The study was approved by the ethical committees of all institutions and was conducted in accordance with the provisions of the Declaration of Helsinki.

HRQoL Assessment for Children

The PedsQL is a validated self-report instrument to measure of HRQoL specifically for pediatric populations. It has been validated in Italian and consists of 23 items in four domains: physical, emotional, social, and school functioning [20, 23]. Scores can be combined to provide an overall total score and also a psychosocial score (emotional, social, and school functioning). Age-specific modules of parent report (ages 2–18) and self-report (ages 8–18 years) were used. Following the authors’ instructions, parents of children age < 8 years completed a proxy-report questionnaire and children aged 8–17 completed a self-report questionnaire with the help of their caregivers. Scores were calculated for each of the four dimensions and for the total and psychosocial summary. Maximum score for each domain is 100 with higher score indicating better the HRQoL [20, 23]. Mean scores for patients managed conservatively were compared with published norms for healthy children and for patients living with chronic diseases, including diabetes mellitus type 1 (DMT1), rheumatology diseases (RD), and malignancies [24,25,26]. PEDs scores for post-HSCT and patients managed conservatively were also compared.

The second tool used, the SDQ, is a validated instrument for screening psychopathology in children and adolescents [21]. Age-specific modules of parent report (ages ≤ 4 and ≥ 5) and self-report (ages > 11) were used. The SQD consists of five scales: emotional difficulties, conduct disorders, hyperactivity, peer difficulties, and the prosocial category. A total score is generated for all domains except for the prosocial score. Higher scores indicate more difficulties. For the prosocial domain, the best score possible is 10. Scores recorded for participants managed conservatively and for those who underwent HSCT were compared with published norms for healthy children of UK population [21] due to the lack of published Italian normative data.

HRQoL Assessment for Adults

Adults completed the SF-12, a 12-items questionnaire specifically validated to measure HRQoL in the population over 18 years of age. Scores were calculated for two domains: mental component summary (MCS) and physical component summary (PCS). Each component is scored on a scale from 0 to 100, with higher scores representing better health [27]. Mean scores for CGD participants treated conservatively were compared with published norms for healthy Italian adults (www.istat.it) and with published data on chronic conditions: DMT1, RD, and malignancies [27, 28] (https://www.istat.it/it/files//2017/12/Bes_2017.pdf). Data from CGD adults managed conservatively were also compared with data from CGD adults who underwent HSCT.

Statistical Analysis

Continuous data were described by mean and standard deviation (SD). Categorical data were presented as counts and percentages. Comparisons of continuous parameters between treatment groups were calculated with a t test if normally distributed and with a Mann-Whitney U test if not normally distributed; differences in frequencies between groups were calculated by chi-square exact test. Correlation between continuous variables were computed using Spearman’s rank correlation coefficient. Statistical significance was set at p > 0.05. Data were processed and analyzed using the SPSS version 25.

Results

Patients’ Characteristics

A total of 47 patients (28 adults and 19 children) were enrolled in the study. The sociodemographic characteristics of patients and caregivers included in the analysis are summarized in Table 2. Overall, the mean age was 7.3 ± 4.4 years for pediatric patients and 27.6 ± 8.0 years for adults. X-linked was the most frequently occurring mode of inheritance (83%). The mean age at diagnosis was 1.7 ± 1.7 years for pediatric patients and 7.3 ± 10.6 year for adults (p = 0.028). Clinicians reported a good/very good clinical conditions in 32 (68%), a poor condition in 13 (28%), and a severe condition in two patients (4%). Eight children and three adults underwent HSCT at least 1 year before the enrollment: the mean time post-HSCT was 18.5 ± 6.8 months for adults and 38.0 ± 25.5 months for children. All not-HSCT patients received antimicrobial prophylaxis with antifungal agents. No participants received prophylactic interferon gamma.

Table 2 Characteristics of CGD participants and caregivers of pediatric patients
Full size table

HRQoL Assessment in Children

Seventeen proxy and 8 patients reports were completed in CGD pediatric group. Figure 1a describes the result of the comparison between non-HSCT CGD children, healthy controls, and patients affected by chronic diseases for PEDsQoL scores as reported by caregivers. Among the CGD non-HSCT group, proxies reported lowest mean scores in the items related to school functioning (68.3 ± 23.9) and highest mean scores in those related to physical functioning (90.4 ± 5.5). In comparison with healthy peers, children with CGD had higher impairment in social (91.4 ± 14.4 vs 83.0 ± 12.3, p = 0.05) and school functioning (85.5 ± 17.6 vs 68.3 ± 23.9, p = 0.05). When compared with children with chronic diseases, CGD patients had less disabilities (total score) than children with RD, cancer, and DMT1 (83.4 ± 8.7 vs 71.0 ± 18.5, p = 0.0012, 69.7 ± 19.2, p = 0.0006, and 76.6 ± 14.3, p = 0.040, respectively). They also were less affected in the physical (90.4 ± 5.4 vs 66.7 ± 24.1, p < 0.0001; 68.7 ± 25.0, p < 0.0001; 82.3 ± 18.0, p = 0.0012, respectively) and emotional dimensions (85.5 ± 10.5 vs 69.0 ± 21.4, p = 0.0007; 67.5 ± 20.3, p = 0.0004; 68.5 ± 19.3, p = 0.0005, respectively). Similar scores were recorded in social and school areas between CGD and chronic diseases.

Fig. 1
figure 1

Proxy (a) and patient (b) PedsQL scales scores of pediatric patients with CGD. Scores from non-HSCT CGD participants are compared with children with other chronic diseases and the healthy pediatric population. CGD Chronic granulomatous disease, RD rheumatic diseases, DMTI diabetes mellitus type 1. *P ≤ 0.05; **P ≤ 0.01; ****P ≤ 0.001; ****P ≤ 0.0001

Full size image

In addition, participants ≥ 6 years old were reported to have higher impairment in proxy-PEDsQL total score, psychosocial summary and social functioning dimension in comparison with those aged < 6 years (75.8 ± 11.3 vs 90.2 ± 5.4, p = 0.003, 73.8 ± 10.5 vs 88.6 ± 7.5, p = 0.004, and 77.1 ± 7.6 vs 93.5 ± 10.3, p = 0.003, respectively) (Table S1). Caregivers also described higher impairment in total score, psychosocial summary, and school functioning for children having poor/severe health status in comparison with those with good/very good status (72.9 ± 13.0 vs 87.8 ± 7.6, p = 0.011, 71.2 ± 9.8 vs 86.0 ± 9.7, p = 0.018, and 57.5 ± 29.0 vs 82.5 ± 14.7, p = 0.049, respectively). Mother and father’s education level and occupation did not influence the PEDSsQL scores (Table S1).

Differently from caregivers, patients reported the lowest scores on emotional and school dimensions (73.0 ± 28.4 and 73.0 ± 19.2, respectively) and the highest scores in the social functioning area (76.0 ± 22.7, Fig.1b). Data of self-report scores of CGD children were not informative because of the small sample size of not-HSCT transplanted patients who completed the PEDsQL (Fig. 1b and Table S2).

No significative difference was recorded by proxy- and patient- reported PEDsQL scores in patients who underwent HSCT and those who were managed conservatively. However, for HSCT recipients, we recorded not-significant improvement in total, psychosocial summary, social, and school functioning scores by proxy-reports (Fig. 2a), and in social and school functioning scores by self-reports (Fig. 2b).

Fig. 2
figure 2

Proxy (a) and patient (b) PedsQL scales scores of non-HSCT (white histograms) and post-HSCT children (gray histograms) with CGD. Discontinued lines represent the mean value for normative population

Full size image

Nineteen children’s parents and three patients > 11 years completed the SDQs. Eight children were post-HSCT (Table 3). As reported by both proxies and patients, children treated conservatively and those who underwent HSCT had significantly higher peer problems scores when compared with population norms. Moreover, non-HSCT and post-HSCT participants aged ≥ 5 years showed higher difficulties in proxy- and self-reported total scores in comparison with age-matched healthy children. No significant difference was observed between patients treated with HSCT and those treated conservatively (data not shown).

Table 3 SDQ report for patients aged ≤ 4 years and for patients ≥5 years (proxy report; children report)
Full size table

HRQoL Assessment in Adult Patients

Overall, patients with CGD showed a decreased well-being, compared with the general Italian population (PCS score 47.8 ± 8.8 vs 53.7 ± 5.9, p < 0.0001; MCS score 45.3 ± 12.5 vs 52.8 ± 7.9, p < 0.0001, Fig. 3a).

Fig. 3
figure 3

SF-12 PCS and MCS scales scores of adults with CGD. Scores from non-HSCT CGD participants are compared with patients with other chronic conditions and the healthy adult Italian population (a). Scores from non-HSCT CGD (white histograms) were compared with post-HSCT CGD adults (gray histograms) (b). CGD Chronic granulomatous disease, RD rheumatic diseases, DMTI diabetes mellitus type 1, PCS physical component summary, MCS mental component summary. *P ≤ 0.05; **P ≤ 0.01; ****P ≤ 0.001; ****P ≤ 0.000

Full size image

When compared with adults with chronic diseases, PCS scores were higher for non-HSCT CGD than RD, malignancies, and DMT1 (33.4 ± 9.1, p < 0.0001, 40.1 ± 10.0, p = 0.0002, and 40.2 ± 11.9, p = 0.002, respectively). MCS scores were lower in CGD than RD (45.3 ± 12.5 vs 49.9 ± 10.5, p = 0.040) but patients affected by CGD had similar mental impairment (MCS) than those with DMT1 and cancer (Fig.3a). Age, level of education, and working status did not correlate with MCS and PCS scores. Moreover, SF-12 scores did not differ between participants with poor/moderate and good/very good clinical status (Table S3). PCS and MCS scores also did not differ in the non-HSCT and in the post-HSCT group (Fig.3b).

Discussion

In last decades, the overall survival of patients with CGD has significantly improved thanks to prophylactic antimicrobial therapies, early aggressive treatment of infections and the advent of azole antifungals [6, 7], and most of patients now reach adulthood. The aging process of the CGD population poses new challenging problems due to the lifetime recurrence of disease-related symptoms [5, 8, 14]. In the last years, improvements in HSCT techniques for CGD with both matched sibling donors and unrelated donors have led to disease cure [9, 11] and improvement of HRQoL of patients [19]. However, it is still unknown what are the best timing and conditioning HSCT regimen [9, 13]. For these reasons, families often may take the difficult decision to continue on lifelong prophylaxis or to start a curative but potentially life-threatening treatment such as HSCT .

Considering such challenges, it is expected that CGD may affect the psychosocial development of the patient and make the planning of complex support programs necessary. At now, few data are available on HRQoL of children with CGD [15, 19] and to the best of our knowledge no study has been conducted in adults. In a recent series of 63 patients from a single center in the UK, parents and self-reported HRQoL for non-transplanted children with CGD was significantly lower than healthy norms. Emotional difficulties were also reported by proxies in comparison with healthy children [19].

In this multicenter study, we investigated the HRQoL in Italian children and adults followed up in PID care centers belonging to the IPINET group. For children, we used a generic HRQoL instrument, the PEDsQL questionnaire [20, 23] combined with the SDQ, a measure of emotional and behavioral difficulties [21)]. For adults, we used the SF-12 questionnaire, a HRQoL measure including dimensions on physical and mental aspects of well-being [22]. Here we did not confirm the observation from the UK cohort, showing that children with CGD did not show a poorer quality of life than healthy peers, as rated by parents and children themselves. This data was consistent with similar studies on chronic illnesses in children and adolescents, such as chronic kidney diseases [29]. However, in comparison with healthy, all the age brackets of CGD children were shown to be at increased risk of difficulties with peers, potentially exposing patients to psychological problems and emotional difficulties. Differently, adults with CGD showed a decreased well-being when compared with the general population, both in physical than in mental dimensions, possibly due to the incoming comorbidities over the time [5, 8, 14]. Moreover, for children, we observed a better overall parents-reported HRQoL in CGD in comparison with peers living with RD, DMT1, and cancer. Similarly, in adults, we observed less impairment in physical functions in CGD than in RD, malignancies, and DMT1. However, adults with CGD were similarly affected in mental dimension in comparison with those affected by chronic diseases. Children aged ≥ 6 years and those compromised/poor health status had worse overall HRQoL and psychosocial activity, as reported by caregivers. Poor health condition had also a negative impact on the school functioning as scored by parents. Differently, in adults, we did not identify clinical or demographic characteristics with a negative impact on QoL.

HSCT has previously been shown to have a positive impact upon QoL of children with CGD in comparison with those managed conservatively [19]. In this study, we could not confirm this observation neither in pediatric nor in adult cohort, due to the small sample of participants who underwent HSCT for each age range. Another concern is that the time of evaluation post-HSCT was < 24 months for most participants, a time where patients may still be experiencing post procedure complications. Longitudinal studies evaluating HRQoL before and after HSCT treatment are desirable to clarify the role of HSCT in reducing disabilities in CGD.

Also for CGD patients treated conservatively, the main drawback of the study is the small number of subjects enrolled in each age group. However, as reported by Apolone et al., the number of adults enrolled can be considered enough to provide reliable statistics [30].

By definition, HRQoL is the subjective perception of health status concerning physical, psychological, social functioning, and well-being. It is well documented that clinicians’ evaluation of HRQoL does not fit well with ratings made by patients themselves [31, 32]. Actually, in clinical practice, patient-reported outcome measures may be a means of facilitating doctor-patient communication, uncovering patients’ problems, monitoring disease or treatment, and screening for functional problems [16, 33]. For these reasons, their use should be encouraged. Our results highlight the importance to extend the assessment of CGD from simple clinical disease-parameters to patient-reported HRQoL measures. As for other chronic diseases [34, 35], our data can contribute to the debate about how the patients with CGD and their caregivers perceive the burden of disease across different life domains, allowing to improve care strategies and their quality of life.