You are viewing the site in preview mode

Skip to main content
Download PDF
  • Research article
  • Open access
  • Published:

Mortality in children and adolescents with autoimmune inflammatory rheumatic diseases admitted to the pediatric intensive care unit

Pediatric Rheumatology volume 23, Article number: 20 (2025) Cite this article

Abstract

Background

This study aimed to describe the characteristics and outcomes of children and adolescents with autoimmune inflammatory rheumatic diseases (AIIRD) who were admitted to the pediatric intensive care unit (PICU). The accuracy of the Pediatric Risk of Mortality (PRISM) III and Pediatric Index of Mortality (PIM) 3 scores to predict the mortality were investigated.

Methods

This was a retrospective cohort study. Children and adolescents with AIIRD aged ≤ 18 years who were admitted to the PICU at the largest university-based referral center in Thailand during July 2011 to June 2021 were included.

Results

There were 122 PICU admissions from 74 patients; mean age of 12.0 ± 4.3 years, 74.3% female. Majority of AIIRD were systemic lupus erythematosus (SLE) (83.8%), followed by systemic juvenile idiopathic arthritis (5.4%), juvenile dermatomyositis (JDM) (2.7%) and microscopic polyangiitis (2.7%). The main cause of admission was combined infection and disease flare (29.5%). Pneumonia was the main site of infection. Acinetobacter baumanii was the most common causative agent. Macrophage activation syndrome occurred in 8 (6.5%) admissions. The mortality rate of PICU admissions was 14.8% from 18 deaths; 17 with SLE and 1 with JDM. Mechanical ventilation (aOR 24.07, 95%CI:1.33-434.91, P= 0.031), pneumothorax (aOR 24.08, 95%CI:1.76-328.86, P = 0.017 and thrombocytopenia (aOR 8.34, 95%CI:1.31–53.73, P = 0.025) were associated with mortality. The risk of mortality rate as predicted by the PRISM III score increased with a score ≥ 9. For the PIM 3 score, the risk of mortality increased if the score ≥ 3. The area under the ROC curve for the PRISM III and PIM 3 scores was 0.741 (95%CI: 0.633–0.849), P = 0.001 and 0.804 (95%CI: 0.685–0.924), P < 0.001, respectively. The model calibration using the Hosmer-Lemeshow goodness of fit test demonstrated a chi-square of 4.335, P = 0.826 for PRISM III and 7.987, P = 0.435 for PIM 3.

Conclusion

SLE was the main AIIRD that required admission to the PICU. Mechanical ventilation, pneumothorax and thrombocytopenia were associated with mortality in pediatric patients with AIIRD. The PRISM III and PIM 3 scores demonstrated good calibration, while the PIM 3 score provided better discrimination ability in the prediction of mortality for pediatric AIIRD.

Background

Autoimmune inflammatory rheumatic diseases (AIIRD) comprise various chronic inflammatory diseases affecting multiple structures of the skeletal system, blood vessels, and internal organs [1]. AIIRD can be classified into autoimmune diseases such as systemic lupus erythematosus (SLE), juvenile dermatomyositis (JDM), mixed connective tissue diseases, vasculitis and autoinflammatory diseases. Due to the natural course of the disease, patients with AIIRD may experience flare-ups of the disease [2, 3]. Additionally, patients with AIIRD may encounter complications from the disease or adverse reactions from immunomodulatory therapy. Immunomodulatory treatment of patients with AIIRD consists of mainly glucocorticoids and/or immunosuppressive drugs including methotrexate, azathioprine, cyclophosphamide, mycofenolate mofetil and biologics. The treat-to-target (T2T) approach was proposed in AIIRD and remission is the ideal treatment target [4,5,6]. Low disease activity is an alternative treatment target in patients with longstanding disease [6,7,8]. In childhood-onset SLE, Lupus Low Disease Activity State (LLDAS) was associated with lower damage accrual and a reduction in severe flares [9, 10]. Severe flares and complications require comprehensive care in the intensive care unit [11,12,13,14].

Up to one-third of hospitalized patients with AIIRD required admission to the pediatric intensive care unit (PICU) [15]. The main indications of PICU admissions may include life-threatening manifestations of AIIRD, infections, flare-ups of diseases, disease complications, or causes not related to AIIRD [11, 13, 15, 16]. Mortality rates of patients with AIIRD were reported between 9.13 and 15% [11, 13, 17, 18]. Predictors of mortality included the use of invasive mechanical ventilation, renal replacement therapy, vasoactive-inotropic agents, and cardiac failure [13, 19].

Currently, several scoring systems have been proposed to predict the outcome in critically ill patients [20]. The most widely used scores in pediatric patients include the Pediatric Risk of Mortality (PRISM) III scores [21] and the Pediatric Index of Mortality (PIM) 3 [22]. The utility of both scores varied among studies with conflicting results [20, 23, 24]. Since the data relating to the outcome of admissions to the PICU among pediatric AIIRD are rare, particularly in the countries of South East Asia. Additionally, to our knowledge, there are no studies focusing on the utility of the PRISM III and PIM 3 scores in predicting AIIRD mortality. Therefore, our objectives in this study were to describe the characteristics, outcomes of PICU admissions and factors associated with the mortality of pediatric patients with AIIRD. The utility of the PRISM III and PIM 3 scores to predict mortality was investigated.

Methods

Study design, setting and population

This retrospective cohort study was conducted at the Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. Siriraj Hospital is Thailand’s largest university-based national tertiary referral center with 303 beds for pediatric inpatient admissions. There were 12 beds in the PICU, excluding the cardiac intensive care unit and the neonatal intensive care unit. Data from AIIRD patients aged ≤ 18 years who were admitted to the PICU between July 2011 and June 2021 from electronic medical records were collected. AIIRD included systemic lupus erythematosus (SLE), juvenile dermatomyositis (JDM), systemic scleroderma, mixed connective tissue disease, sjögren syndrome, juvenile idiopathic arthritis (JIA), granulomatosis with polyangiitis, microscopic polyangiitis, eosinophilic granulomatosis with polyangiitis, takayasu arteritis, kawasaki disease, polyarteritis nodosa, henoch-schönlein purpura (IgA vasculitis), sarcoidosis and behçet disease. The protocol for this study received ethical approval from the Siriraj Institutional Review Board (SIRB) (COA no. Si 416/2021).

Data collection and outcomes

The data collected included demographic data, type of AIIRD, disease activity, cause of admission to the PICU, organ involvement, comorbidity, length of stay, intervention, and treatment received in the PICU. Each admission to the PICU was evaluated to explore the study outcomes. The outcome of admission to the PICU was classified as survivors and non-survivors. Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) score [25] was used to assess disease activity in patients with SLE. Macrophage activation syndrome (MAS) was defined based on the 2016 classification criteria for MAS in systemic JIA [26] and the preliminary diagnostic guidelines for MAS in SLE [27].

Mortality prediction scores

The PRISM III and PIM 3 scores were calculated [21, 22]. The PRISM III score is calculated using clinical and laboratory variables collected during the first 12 h of admission to the PICU as follows; Glasgow Coma Score, temperature, respiratory rate, systolic and diastolic blood pressure, heart rate, pupillary reactions, arterial blood gas, glucose, urea, creatinine, potassium, white blood cell count, platelets, prothrombin time and partial thromboplastin time [21]. The PIM 3 score is calculated using variables collected from initial contact with the patient to 1 h after arrival in the PICU, including diagnosis, systolic blood pressure, base excess, type of admission, FiO2, PaO2, mechanical ventilator support, recovery from surgery as indication of admission to the PICU, admission to cardiac bypass and pupillary reactions [22].

Statistical analysis

All data analyses were performed using the SPSS Statistics version 22.0 (SPSS, Inc., Chicago, IL, USA). The sample size was calculated using the proportion of mortality by Radhakrishna et al. [13] at 15% by the formula n = Z2ɑ/2P(1-P)/d2 (ɑ = type I error = 0.05, 2-sided, 95% CI, Z = 1.96; P = 0.15; d = 0.07), which results in a calculated sample size for at least 100 admissions. Due to the potentially limited number of patients and events, all eligible participants and events were recruited during the study period. Descriptive statistics were used to summarize demographic and clinical characteristics data, and the results of those comparisons were given as numbers and percentages. Continuous data were shown as mean ± standard deviation (SD) for normally distributed data and median and interquartile range (IQR) for non-normally distributed data. Continuous data comparisons were performed using Student’s t-test or Mann-Whitney U test. Comparisons of categorical data were performed using either the chi-square test or Fisher’s exact test. Multivariable logistic regression analysis was used to identify the factors associated with mortality. Using univariable logistic regression analysis, variables associated with mortality with P-value < 0.05 were entered into multivariable logistic regression analysis using the backward method. The results were reported as odds ratio (OR) and 95% confidence interval (CI). Variables with a P-value less than 0.05 were considered statistically significant.

To evaluate the utility of the PRISM III and PIM 3 scores to distinguish between nonsurvivors and survivors, the receiver operating characteristic (ROC) curve was plotted against the outcome for the two scores. Discrimination, or the accuracy of a model in differentiating outcome groups, was assessed by the area under the ROC curve. This area under the curve is an expression of the overall discrimination across the range of risks and is a good summary measure of the predictive ability. The cut-off points extracted from the ROC curve were calculated with 95%CI. The model calibration was assessed using the Hosmer-Lemeshow goodness of fit test, where acceptable calibration was defined by the P-value ≥ 0.05, meaning that there was no significant difference between the predicted mortality by the score and the observed mortality for the study population. The standard mortality ratio (SMR) was calculated by dividing the observed deaths by cumulative expected deaths, a ratio greater than 1.0 indicates that prediction is under, while a ratio less than 1.0 indicates prediction is over.

Results

Demographic data and clinical characteristics

A total of 122 PICU admissions from 74 patients (female 74.3%) with a mean age at the time of AIIRD diagnosis of 10.7 ± 3.8 years. Mean age at the time of PICU admission was 12.0 ± 4.3 years. The majority of AIIRDs were SLE (83.8%), followed by systemic JIA (5.4%) and JDM (2.7%) and microscopic polyangiitis (2.7%). The first diagnosis of AIIRD was made during admission to the PICU in 20 (27%) patients. Demographic data and clinical characteristics of 74 patients with AIIRD are shown in Table 1. The clinical characteristics and organ involvement during admission to the PICU classified by AIIRD are shown in the Supplementary Table 1.

Table 1 Demographic data of patients with autoimmune inflammatory rheumatic diseases (N = 74)
Full size table

Causes of PICU admission

Of 122 admissions, the main cause of admission to PICU was co-existing infection and disease flare-up at 36 (29.5%) admissions. Disease flare-up as the cause of admission to PICU found in 32 (26.2%) admissions. Infection as the only reason for admission to the PICU identified in 23 (18.9%) admissions. The complications of AIIRD as an indication for admission to the PICU occurred in 14 (11.5%) admission; six from posterior reversible encephalopathy syndrome, three from volume overload, two from pulmonary embolism, two from gastrointestinal bleeding, and one from intracranial hemorrhage. Postoperative care and anaphylaxis were documented in 14 (11.5%) and 3 (2.5%) admissions, respectively.

Pneumonia was observed in 56 (45.9%) admissions. Acinetobacter baumannii and Stenotrophomonas maltophilia were the most common causative agents of bacterial pneumonia at seven (25%) admissions each. Invasive pulmonary aspergillosis was found in 20 (37.5%) admissions. Sepsis was found in 33 (27%) admissions. Acinetobacter baumannii was the main causative agent in eight (29.6%) admissions, followed by Salmonella spp. in three (11.7%) admissions. Fungemia occurred in two admissions due to Candida tropicalis and Cryptococcus neoforman. Urinary tract infection was detected in 15 (12.3%) admissions; most were Candida spp. (9 admissions; 60%). The details of the causative agents are shown in the Supplementary Table 2.

Of the 14 SLE PICU admissions with pneumothorax, the pneumothorax coexisted with pneumonia in all. The most common causative agent in pneumothorax complicated pneumonia was Aspergillus spp. (9/14 = 64.3%). There were CMV (2/14, 14.2%), Nocardia spp. (2/14, 14.2%), Salmonella group E (2/14, 14.2%), and Stenotrophomonas maltophilia (2/14, 14.2%) pneumonia with pneumothorax. The rest of the causative agents of pneumonia with pneumothorax included Acinetobacter baumannii, non-fermenting gram-negative bacilli and Pseudomonas aeruginosa each admission. Of the 14 SLE PICU admissions with pneumothorax, 7 (50%) of them required mechanical ventilator support; 2 with non-invasive mechanical ventilators (BIPAP) and 5 patients with invasive mechanical ventilators. One patient additionally required the ECMO support.

Outcome of PICU admission and prognostic factors

The mortality rate for admission to the PICU was 14.8%. Nonsurvivors occurred in 18 patients of 122 PICU admissions; 17 of them were SLE, 1 with JDM. Of 17 non-survivors with SLE, seven of them were attributed to infections, five with a combination of infection and disease flare-up, one with severe lupus myocarditis and cardiogenic shock, and four with disease complications. The patient with JDM with melanoma differentiation-associated protein (MDA) 5 associated with rapidly progressive interstitial lung disease died of acute respiratory distress syndrome complicated by pneumonia. The detailed of cause of nonsurvivors is shown in the Table 2.

Table 2 Causes of mortality of nonsurvivors with AIIRD
Full size table

Of 122 admissions, the median length of stay in the PICU was 6 days (IQR 3-14.25) days. Macrophage activation syndrome occurred in 8 (6.5%) admissions. Multiple organ failure was present in 31 (25.4%) admissions. The renal system was the most common organ involvement (80.3%), followed by the respiratory system (79.5%) and the cardiovascular system (53.3%).

Invasive mechanical ventilator was used in 48 (39.3%) admissions. Cardiopulmonary resuscitation was performed in 11 (9%) admissions. Plasmapheresis and continuous renal replacement therapy were applied in 22 (18%) and 21 (17.2%) admissions, respectively. Extracorporeal membrane oxygenation was used in 3 (2.5%) admissions. Vasoactive inotropic agents were administered in 37 (30.3%) admissions. IV pulse methylprednisolone was treated in 35/122 (28.7%) admissions. Receiving IV pulse methylprednisolone was not significantly different between survivors (30/104 = 28.8%) and nonsurvivors (5/18 = 27.8%), P = 0.926. The median daily prednisolone dosage was 45 (IQR 17.5–60) mg. There was no significant difference of the median daily prednisolone dosage between survivors (45 mg, IQR 15–60) and nonsurvivors (42.5 mg, IQR 30–60), P = 0.920. The PRISM III and PIM 3 scores were significantly higher in non-survivors than survivors at 9.5 (5.8–12) versus 5 (3-8.8), P = 0.001 and 4.6 (2.5–9.9) versus 1.3 (0.4–1.9), P < 0.001, respectively (Fig. 1). Comparison of variables between survivors and nonsurvivors is shown in supplementary Table 3 (clinical characteristics, comorbidities and prior treatments), supplementary Table 4 (organ involvements, PRISM III and PIM 3 scores) and supplementary Table 5 (interventions and treatments). Multivariable logistic regression analysis found that the mechanical ventilator used (aOR 24.07, 95%CI: 1.33-434.91, P = 0.031), pneumothorax (aOR 24.08, 95%CI:1.76-328.86, P = 0.017 and thrombocytopenia (aOR 8.34, 95%CI: 1.31–53.73, P = 0.025) were associated with mortality (Table 3).

Fig. 1
figure 1

Box plot graphs of the PRISM III (A) and PIM 3 (B) scores compare between survivors and nonsurvivors of all admissions

Full size image
Table 3 Univariable and multivariable logistic regression analysis for factors associated with the mortality
Full size table

The discriminative performances of PRISM III and PIM 3 and the mortality are shown in Fig. 2. The area under the ROC curve for the PRISM III and PIM 3 scores to predict mortality was 0.741 (95%CI: 0.633–0.849), P = 0.001 and 0.804 (95%CI: 0.685–0.924), P < 0.001, respectively. The risk of mortality rate as predicted by the PRISM III score increased with a score ≥ 9. The observed mortality rate in PRISM III cases of less than 9 was 7.1% compared to 31.6% in cases with higher scores (P-value = 0.001). For the PIM 3 score, the risk of mortality increased if the score ≥ 3. The observed mortality rate in case of PIM 3 less than 3 was 5.6% compared to 39.4% in case of higher score (P-value < 0.001). The PRISM III score at the cut-off point of 9 had a sensitivity of 66.7% (95% CI 40.9–86.6) and specificity of 75% (95%CI: 65.5–82.9) while the cut-off point of 3 of the PIM 3 score to predict mortality had a sensitivity of 72.2% (95%CI: 46.5–90.3) and a specificity of 80.8% (95%CI:71.8–87.8).

Fig. 2
figure 2

Area under the receiver operating curve (ROC) of PRISM III and PIM 3 scores to predict mortality

Full size image

The model calibration using the Hosmer-Lemeshow goodness of fit test demonstrated a chi-square of 4.335, P = 0.826 for PRISM III and 7.987, P = 0.435 for PIM 3. The standardized mortality ratio (SMR) of the PRISM III and PIM 3 were 4.5 (95% CI: 2.75–6.97) and 3.6 (95% CI: 2.20–5.57), respectively. The performance characteristics of the PRISM III and PIM 3 scores are shown in Table 4.

Table 4 Performance characteristics of PRISM III and PIM 3 scoring systems in predicting mortality of patients with AIIRD admitted to the PICU
Full size table

Discussion

Our study demonstrated that SLE was the main AIIRD that required admission to the PICU and the combination of infection and flare-up of underlying AIIRD was the main cause of admission. The mortality rate for admission to the PICU was 14.8% and almost all nonsurvivors were patients with SLE. The occurrence of invasive mechanical ventilator used, pneumothorax and thrombocytopenia were predictors of mortality. The PRISM III and PIM 3 scores demonstrated good calibration, while the PIM 3 score provided better discrimination ability in the prediction of mortality for pediatric AIIRD.

Coexisting infection and disease flare-up was the most common cause of patients to be admitted to the PICU in our AIIRD cohort. Immunocompromised patients with AIIRD are at increased risk of developing severe or opportunistic infections [28, 29]. Immune dysfunction, inadequate response to vaccinations, and previous organ damage in AIIRD were all at risk of severe infections [14, 30]. Infection could also trigger a disease flare-up [31]. Active AIIRD treated with immunomodulatory therapy and secondary infection could result in a poor outcome [14]. The mortality rate of admission to the PICU was 15% in a large tertiary care center in the United States between 1995 and 2009 and the deaths were attributed to coexisting infection and disease flare-up [13]. Contrary to our findings, disease flare-up or disease-related complication was the main cause of admission to the PICU demonstrated by Al-Mayouf et al. [11]. Radhakrishna et al. reported that AIIRD-related complication was the main reason for admission to the PICU up to 50% [13]. Our study determines that not only the active underlying AIIRD, but also infection, were major concerns leading to admission to the PICU. Therefore, achieving remission or low disease activity of AIIRD and prompt appropriate infection treatment are essential to reducing morbidity and mortality in patients with AIIRD.

Pneumonia was the predominant infection of AIIRD from our observation. Acinetobacter baumannii and Stenotrophomonas maltophilia were the 2 most common causative agents of bacterial pneumonia. Invasive pulmonary aspergillosis was another major infection in our study. A previous study similarly found that pneumonia was found to be the leading cause of admission to the ICU of adults with SLE [32] and Acinetobacter baumannii was the organism most identified in SLE patients [32, 33]. Another study focusing on children with lupus nephritis found that infection-related mortality occurred 15.1% and invasive fungal infection was the predominant reason [29].

In our study, SLE was the main cause of PICU admission in patients with AIIRD. Similarly, SLE was the main AIIRD that required ICU care [11, 13, 19]. Childhood-onset SLE (cSLE) had more major organ involvement, disease severity, and mortality than adult SLE [34]. The SMR of cSLE was high at 18.8 [34], while the SMR of adults with SLE ranged lower at 2.2–2.9 [34, 35]. In a report by Jongvilaikasem et al. the mortality rate was 3.3 per 100 years in Thai c-SLE patients [36]. Although SLE-related disease was the main cause of death in a study by Joo and colleagues [34], in our investigation infection was the leading cause of death in cSLE.

Apart from SLE, systemic JIA and JDM were the subsequent AIIRD that required admission to the PICU in our cohort. Macrophage activation syndrome (MAS) was the key complication of systemic JIA and may result in high mortality [37]. In a study by Al-Mayouf et al., MAS complicated by sepsis was found to be the main cause of death in children with AIIRD [11]. Our JDM patient who died had MDA-5 associated rapidly interstitial lung disease who developed acute respiratory distress syndrome and air leak syndrome complicated by pneumonia. Acute respiratory failure was the main reason for admission to the ICU in patients with idiopathic inflammatory myopathy [38] similar to our reported patient. As MDA-5 dermatomyositis carries a poor prognostic outcome, early detection and aggressive immunomodulatory treatments should be emphasized to minimize mortality [39].

We observed that the use of an invasive mechanical ventilator and pneumothorax were significantly associated with mortality. Mechanical ventilation, renal replacement therapy, and vasoactive inotropic agents were found to be predictors of mortality in AIIRD patients [13]. In addition, invasive mechanical ventilation was one of the factors associated with mortality in the ICU in adults with AIIRD [19]. Interestingly, pneumothorax was documented up to 15 admissions in our study (14 with SLE and 1 with JDM) and developed in 33% of nonsurvivors. Pneumothorax could be secondary to barotrauma from mechanical ventilation, ARDS, necrotizing pneumonia, or interstitial lung disease. Noticeably, complications or failure of the respiratory system are significant predictors of mortality.

Scoring systems may predict the mortality risk in critically ill patients. In pediatric populations, the PRISM III and PIM 3 scores were the most widely used [23, 24]. The PRISM III and PIM 3 scores performed well in predicting mortality in PICU [20]. Rahmatinejad et al. described the better performance of PRISM III than PIM 3 to predict mortality in patients admitted to the PICU [23]. Alkhalifah et al. also reported a better discrimination performance of PRISM III than of PIM 3 in patients with central nervous system and metabolic/genetic diseases [24]. The meta-analysis found that the PRISM-III / IV scores had an SROC of 0.84 (95% CI: 0.80–0.87), while PIM-3 had an SROC of 0.82 (95% CI 0.78–0.85) [20]. It is noteworthy that the area under the ROC curve for the PRISM III and PIM 3 scores to predict mortality in our study was 0.741 (95%CI: 0.633–0.849), P = 0.001 and 0.804 (95%CI: 0.685–0.924), P < 0.001, respectively. Our study therefore demonstrated that the PIM 3 score had better discrimination performance than the PRISM III score in patients with AIIRD. The varied discrimination ability of the PRISM III and PIM 3 scores may be secondary to the different time points of evaluating variables during admissions to the PICU. Additionally, the PIM 3 score is more feasible than PRISM III in terms of the number of variables and the time to collect the data. Although poor calibration of the PRISM III and PIM 3 scores was described in previous studies [23, 24], our study determined good calibration of both scores. The application of scoring systems to predict mortality in pediatric AIIRD should be a focus of further study.

The mortality rate in patients with AIIRD was relatively high at 14.8%. A possible explanation is that patients with AIIRD are complex given their complicated disease pathogenesis, multifaceted presentations, and unpredictable natural course of diseases. Thus, to alleviate mortality from admission to the PICU in AIIRD, several strategies should be stressed. Achieving remission or low disease activity should be the target of AIIRD treatment [4, 6, 7]. Personal hygiene care, complete vaccination and prophylaxis of opportunistic infections in immunosuppressed patients with AIIRD should be encouraged [28]. Moreover, early detection and prompt intervention in impending critical situations in patients with severe AIIRD are necessary [40]. Our center may be representative of PICU care in a low-resource setting. There is room for improvement in several aspects including manpower, quality of care, and innovative equipment.

In terms of limitations, our data were retrieved from a single center. The retrospective study design could contribute to potential bias, confounders, and missing data. However, we believe that the chance of missing important data is small, since the clinical information in our center was collected in the electronic medical records. Another point is that due to the 10-year period of study data, the trend of treatment practices may change over time and could impact to the outcome of treatment.

Conclusions

The mortality rate was high among pediatric patients with AIIRD in particular SLE. The combination of infection and active underlying AIIRD was the major cause of PICU admissions. To our knowledge, our study is the first study to report the utility of PRISM III and PIM 3 scores to predict mortality in pediatric AIIRD. Both the PRISM III and PIM 3 scores demonstrated good calibration and the PIM 3 score provided better discrimination ability. Treat-to-target strategy to achieve remission or low disease activity of underlying AIIRD in addition to the appropriate infectious prevention and treatment should be underlined. Optimal immunomodulatory treatments in patients with flare-up of underlying AIIRD and severe infection merit further study. Early detection and prompt intervention in critically ill patients with AIIRD by the multidisciplinary team are essential. A prospective study of the application of scoring systems to predict mortality in pediatric AIIRD patients is warranted.

Data availability

The datasets used and/or analysed during the current study are available upon reasonable request.

Abbreviations

AIIRD:

Autoimmune inflammatory rheumatic diseases

IgA:

Immunoglobulin A

JDM:

Juvenile dermatomyositis

JIA:

Juvenile idiopathic arthritis

LLDAS:

Lupus Low Disease Activity State

MAS:

Macrophage activation syndrome

SLE:

Systemic lupus erythematosus

SLEDAI-2K:

Systemic Lupus Erythematosus Disease Activity Index 2000

PICU:

Pediatric intensive care unit

PIM 3:

Pediatric Index of Mortality 3

PRISM III:

Pediatric Risk of Mortality III

ROC:

Receiver operating characteristic

SMR:

Standard mortality ratio

References

  1. Moutsopoulos HM. Autoimmune rheumatic diseases: one or many diseases? J Translational Autoimmun. 2021;4.

  2. Smith EMD, Lythgoe H, Hedrich CM. Current views on lupus in children. Curr Opin Rheumatol. 2022. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/BOR.0000000000000913.

    Article  PubMed  Google Scholar 

  3. Martini A, Lovell DJ, Albani S, Brunner HI, Hyrich KL, Thompson SD, et al. Juvenile idiopathic arthritis. Nat Rev Dis Primers. 2022;8:5.

    Article  PubMed  Google Scholar 

  4. Smith EMD, Aggarwal A, Ainsworth J, Al-Abadi E, Avcin T, Bortey L, et al. Towards development of treat to target (T2T) in childhood-onset systemic lupus erythematosus: PReS-endorsed overarching principles and points-to-consider from an international task force. Ann Rheum Dis. 2023;82:788–98.

    Article  PubMed  Google Scholar 

  5. van Vollenhoven RF, Bertsias G, Doria A, Isenberg D, Morand E, Petri MA et al. 2021 DORIS definition of remission in SLE: final recommendations from an international task force. Lupus Sci Med. 2021;8.

  6. Ravelli A, Consolaro A, Horneff G, Laxer RM, Lovell DJ, Wulffraat NM, et al. Treating juvenile idiopathic arthritis to target: recommendations of an international task force. Ann Rheum Dis. 2018;77:819–28.

    Article  PubMed  Google Scholar 

  7. Smith EMD, Aggarwal A, Ainsworth J, Al-Abadi E, Avcin T, Bortey L, et al. PReS-endorsed international childhood lupus T2T task force definition of childhood lupus low disease activity state (cLLDAS). Clin Immunol. 2023;250:109296.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Franklyn K, Lau CS, Navarra SV, Louthrenoo W, Lateef A, Hamijoyo L, et al. Definition and initial validation of a Lupus Low Disease Activity State (LLDAS). Ann Rheum Dis. 2016;75:1615–21.

    Article  CAS  PubMed  Google Scholar 

  9. Na Nakorn K, Piyaphanee N, Sukharomana M, Pinpatanapong R, Charuvanij S. Outcomes of achieving lupus low disease activity state and damage accrual in childhood-onset systemic lupus erythematosus. Clin Rheumatol. 2023;42:1655–64.

    Article  PubMed  Google Scholar 

  10. Smith EMD, Tharmaratnam K, Al-Abadi E, Armon K, Bailey K, Brennan M, et al. Attainment of low disease activity and remission targets reduces the risk of severe flare and new damage in childhood lupus. Rheumatology (Oxford). 2022;61:3378–89.

    Article  PubMed  Google Scholar 

  11. Al-Mayouf SM, Fallatah R, Al-Twajery M, Alayed T, Alsonbul A. Outcome of children with systemic rheumatic diseases admitted to pediatric intensive care unit: an experience of a tertiary hospital. Int J Pediatr Adolesc Med. 2019;6:142–5.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Mustafa M, Gladston Chelliah E, Hughes M. Patients with systemic rheumatic diseases admitted to the intensive care unit: what the rheumatologist needs to know. Rheumatol Int. 2018;38:1163–8.

    Article  CAS  PubMed  Google Scholar 

  13. Radhakrishna SM, Reiff AO, Marzan KA, Azen C, Khemani RG, Rubin S, et al. Pediatric rheumatic disease in the intensive care unit: lessons learned from 15 years of experience in a tertiary care pediatric hospital. Pediatr Crit Care Med. 2012;13:e181–6.

    Article  PubMed  Google Scholar 

  14. Shulman AI, Punaro M. Critical care of the pediatric patient with rheumatic disease. Curr Opin Pediatr. 2011;23:263–8.

    Article  PubMed  Google Scholar 

  15. Janssen NM, Karnad DR, Guntupalli KK. Rheumatologic diseases in the intensive care unit: epidemiology, clinical approach, management, and outcome. Crit Care Clin. 2002;18:729–48.

    Article  PubMed  Google Scholar 

  16. Kouchit Y, Morand L, Martis N. Mortality and its risk factors in critically ill patients with connective tissue diseases: a meta-analysis. Eur J Intern Med. 2022;98:83–92.

    Article  PubMed  Google Scholar 

  17. Botan E, Gün E, Şden EK, Yöndem C, Gurbanov A, Balaban B, et al. Characteristics and timing of mortality in children dying in pediatric intensive care: a 5-year experience. Acute Crit Care. 2022;37:644–53.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Qiu J, Lu X, Wang K, Zhu Y, Zuo C, Xiao Z. Comparison of the pediatric risk of mortality, pediatric index of mortality, and pediatric index of mortality 2 models in a pediatric intensive care unit in China: a validation study. Med (Baltim). 2017;96:e6431.

    Article  Google Scholar 

  19. Chabert P, Danjou W, Mezidi M, Berthiller J, Bestion A, Fred AA, et al. Short- and long-term prognosis of acute critically ill patients with systemic rheumatic diseases: a retrospective multicentre study. Med (Baltim). 2021;100:e26164.

    Article  CAS  Google Scholar 

  20. Shen Y, Jiang J. Meta-analysis for the prediction of Mortality Rates in a Pediatric Intensive Care Unit using different scores: PRISM-III/IV, PIM-3, and PELOD-2. Front Pediatr. 2021;9:712276.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Pollack MM, Ruttimann UE, Getson PR. Pediatric risk of mortality (PRISM) score. Crit Care Med. 1988;16:1110–6.

    Article  CAS  PubMed  Google Scholar 

  22. Straney L, Clements A, Parslow RC, Pearson G, Shann F, Alexander J, et al. Paediatric index of mortality 3: an updated model for predicting mortality in pediatric intensive care*. Pediatr Crit Care Med. 2013;14:673–81.

    Article  PubMed  Google Scholar 

  23. Rahmatinejad Z, Rahmatinejad F, Sezavar M, Tohidinezhad F, Abu-Hanna A, Eslami S. Internal validation and evaluation of the predictive performance of models based on the PRISM-3 (Pediatric Risk of Mortality) and PIM-3 (Pediatric Index of Mortality) scoring systems for predicting mortality in Pediatric Intensive Care Units (PICUs). BMC Pediatr. 2022;22:199.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Alkhalifah AS, AlSoqati A, Zahraa J. Performance of Pediatric Risk of Mortality III and Pediatric Index of Mortality III scores in Tertiary Pediatric Intensive Unit in Saudi Arabia. Front Pediatr. 2022;10:926686.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Gladman DD, Ibañez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol. 2002;29:288–91.

    PubMed  Google Scholar 

  26. Ravelli A, Minoia F, Davì S, Horne A, Bovis F, Pistorio A, et al. 2016 classification criteria for macrophage activation syndrome complicating systemic juvenile idiopathic arthritis: a European League Against Rheumatism/American College of Rheumatology/Paediatric Rheumatology International Trials Organisation Collaborative Initiative. Arthritis Rheumatol. 2016;68:566–76.

    Article  PubMed  Google Scholar 

  27. Parodi A, Davì S, Pringe AB, Pistorio A, Ruperto N, Magni-Manzoni S, et al. Macrophage activation syndrome in juvenile systemic lupus erythematosus: a multinational multicenter study of thirty-eight patients. Arthr Rhuem. 2009;60:3388–99.

    Article  CAS  Google Scholar 

  28. Fragoulis GE, Nikiphorou E, Dey M, Zhao SS, Courvoisier DS, Arnaud L et al. 2022 EULAR recommendations for screening and prophylaxis of chronic and opportunistic infections in adults with autoimmune inflammatory rheumatic diseases. Ann Rheum Dis. 2022.

  29. Rianthavorn P, Prurapark P. Risk factors of infection-associated mortality in children with lupus nephritis in under-resourced areas. Lupus. 2019;28:1727–34.

    Article  CAS  PubMed  Google Scholar 

  30. Wang H, Lan L, Chen J, Xiao L, Han F. Peripheral blood T-cell subset and its clinical significance in lupus nephritis patients. Lupus Sci Med. 2022;9.

  31. Johnson D, Jiang W. Infectious diseases, autoantibodies, and autoimmunity. J Autoimmun. 2022:102962.

  32. Siripaitoon B, Lertwises S, Uea-Areewongsa P, Khwannimit B. A study of Thai patients with systemic lupus erythematosus in the medical intensive care unit: epidemiology and predictors of mortality. Lupus. 2015;24:98–106.

    Article  CAS  PubMed  Google Scholar 

  33. Vachvanichsanong P, McNeil E, Dissaneewate P. A 30-year retrospective study on causes of death in childhood-onset systemic lupus erythematosus in a tertiary care centre in Southern Thailand. Clin Experimental Rheumatol. 2019;37:879–84.

    Google Scholar 

  34. Joo YB, Park SY, Won S, Bae SC. Differences in clinical features and mortality between childhood-onset and adult-onset systemic lupus erythematosus: a prospective single-center study. J Rheumatol. 2016;43:1490–7.

    Article  PubMed  Google Scholar 

  35. Tselios K, Gladman DD, Sheane BJ, Su J, Urowitz M. All-cause, cause-specific and age-specific standardised mortality ratios of patients with systemic lupus erythematosus in Ontario, Canada over 43 years (1971–2013). Ann Rheum Dis. 2019;78:802–6.

    Article  PubMed  Google Scholar 

  36. Jongvilaikasem P, McNeil EB, Dissaneewate P, Vachvanichsanong P. Improvement of survival rates in the last decade in Thai childhood-onset systemic lupus erythematosus. Pediatr Rheumatol Online J. 2018;16:62.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Henderson LA, Cron RQ. Macrophage activation syndrome and secondary Hemophagocytic Lymphohistiocytosis in Childhood Inflammatory disorders: diagnosis and management. Paediatr Drugs. 2020;22:29–44.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Peng JM, Du B, Wang Q, Weng L, Hu XY, Wu CY, et al. Dermatomyositis and Polymyositis in the Intensive Care Unit: a single-Center Retrospective Cohort Study of 102 patients. PLoS ONE. 2016;11:e0154441.

    Article  PubMed  PubMed Central  Google Scholar 

  39. McPherson M, Economidou S, Liampas A, Zis P, Parperis K. Management of MDA-5 antibody positive clinically amyopathic dermatomyositis associated interstitial lung disease: a systematic review. Seminars Arthritis Rheum. 2022;53:151959.

    Article  Google Scholar 

  40. Chong SL, Goh MSL, Ong GY, Acworth J, Sultana R, Yao SHW, et al. Do paediatric early warning systems reduce mortality and critical deterioration events among children? A systematic review and meta-analysis. Resusc Plus. 2022;11:100262.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors gratefully acknowledge Ms. Kanokwan Sommai, M.Sc. (Applied Statistics), of the Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, and Ms. Nerisa Thornsri, M.Sc. (Applied Statistics), Division of Clinical Epidemiology, Department of Research, Faculty of Medicine Siriraj Hospital, Mahidol University for assistance with statistical analysis.

Funding

This research did not receive any specific grant from funding agencies in the public commercial, or not-for-profit sectors.

Author information

Authors and Affiliations

  1. Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

    Tinnapat Buranapattama

  2. Division of Pediatric critical care, Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

    Suwannee Phumeetham

  3. Division of Nephrology, Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

    Nuntawan Piyaphanee

  4. Division of Rheumatology, Department of Pediatrics, Faculty of Medicine Siriraj Hospital, Mahidol University, 2 Wanglang Road, Bangkoknoi, Bangkok, 10700, Thailand

    Maynart Sukharomana & Sirirat Charuvanij

Authors
  1. Tinnapat Buranapattama
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Suwannee Phumeetham
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Nuntawan Piyaphanee
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Maynart Sukharomana
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Sirirat Charuvanij
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

TB contributed to concept and design, acquisition of data, analysis and interpretation of data and drafting the manuscript. SP, NP and MS contributed to concept and design, analysis and interpretation of data, summary of results, and critically revising the manuscript for important intellectual content. SC contributed to concept and design, acquisition of data, analysis and interpretation of data, summary of results, drafting and critically revising the manuscript for important intellectual content and is the corresponding author. All authors approved the final version of the manuscript.

Corresponding author

Correspondence to Sirirat Charuvanij.

Ethics declarations

Ethics approval and consent to participate

The protocol for this study received ethical approval from the Siriraj Institutional Review Board (SIRB) (COA no. Si 416/2021).

Consent for publication

Not applicable.

Competing interests

All authors declare no personal or professional conflicts of interest relating to any aspect of this study.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Buranapattama, T., Phumeetham, S., Piyaphanee, N. et al. Mortality in children and adolescents with autoimmune inflammatory rheumatic diseases admitted to the pediatric intensive care unit. Pediatr Rheumatol 23, 20 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12969-025-01068-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12969-025-01068-5

Keywords

Pediatric Rheumatology

ISSN: 1546-0096