Children and adolescents represent 15–20% of all patients with systemic lupus erythematosus (SLE) (1). Although the presentation, clinical symptoms, and findings of laboratory tests in patients with juvenile-onset SLE are similar to those that are seen in adults, SLE that begins in childhood has been considered more severe than adult-onset SLE, with higher rates of organ involvement and a more aggressive clinical course (2). Furthermore, it has been reported that pediatric patients with SLE may need high-dose corticosteroids and immunosuppressive agents for disease control more often than do their adult counterparts (3).
Over the past 2 decades, there has been a marked improvement of prognosis in juvenile-onset SLE, owing to earlier diagnosis and better approaches to treatment (4). As a result, children and adolescents with SLE are living longer and may enter adult life with considerable morbidity, which is secondary to the sequelae of disease activity, side effects of medications, and comorbid conditions. This morbidity may affect their quality of life, raising problems related to physical and psychological adaptation to a chronic illness (5). For this reason, cumulative damage has become recognized as a fundamental independent domain of disease outcome, and it has been increasingly accepted that the optimal management of juvenile-onset SLE should include careful monitoring of the accumulation of irreversible damage in the patient's organ systems. Recently, a number of studies have investigated the long-term damage in patients with juvenile-onset SLE (6–9) using the Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) Damage Index (SDI) (10, 11), which is a measure developed for adult SLE.
However, due to the differences between juvenile- and adult-onset SLE in frequency and severity of disease activity, damage features, and treatment approaches, it cannot be assumed a priori that the clinical measures developed for adults are suitable for use in children and adolescents. Therefore, outcome measures developed for adults need to be subjected to critical evidence-based evaluation of their measurement properties in children and adolescents (12). Furthermore, in juvenile-onset SLE, specific instruments or items are needed, that take into account the effects of the disease on physical- and mental-age–related issues that are inherent to growth and development (13).
Although the SDI has proved to be a valid and reliable instrument in the assessment of patients with juvenile-onset SLE, a number of problems have emerged with its use in pediatric patients (4, 14). First, it does not cover all forms of damage that are seen in children or adolescents with SLE, such as growth failure and delayed puberty. Furthermore, since some SDI items, such as myocardial infarction, pancreatic insufficiency, claudication, gastrointestinal stricture, ruptured tendons, and malignancy, are rarely observed in patients with juvenile-onset SLE, their inclusion in a pediatric damage index has been questioned (14). A peculiar aspect that needs to be taken into account in the assessment of damage in pediatric rheumatic diseases is that children have the ability to recover and regenerate to a greater degree than do adults (15). This issue is not incorporated in the SDI, which covers, by definition, only irreversible damage.
The goal of the present study was to devise a modified version of the SDI for use in children and adolescents with SLE. We investigated the frequency and distribution of organ system damage in a large cohort of patients with juvenile-onset SLE and attempted to identify the sources of damage that are more suitable for inclusion in a pediatric damage index.
- Top of page
- PATIENTS AND METHODS
A total of 1,015 patients from 39 countries were included, as follows: 505 patients from Europe (Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Italy, Latvia, The Netherlands, Northern Ireland, Norway, Poland, Portugal, Serbia and Montenegro, Slovakia, Slovenia, Spain, Sweden, Switzerland, and the UK), 388 patients from North and South America (Argentina, Brazil, Canada, Chile, Costa Rica, Cuba, Mexico, and the US), and 122 patients from Asia and Oceania (India, Israel, Japan, Saudi Arabia, Singapore, South Korea, and Turkey). Patients were evaluated in the study centers between February 1997 and March 2005. Data on each patient were recorded only at 1 time point rather than serially.
The demographic features and the SDI scores of patients in the study cohort are presented in Table 1. Six hundred ten (60.1%) of the 1,015 patients had an SDI score of 0, and 405 patients (39.9%) had a score of ≥1. The mean ± SD SDI score was 0.8 ± 1.4 (median 0, range 0–12). The SDI score increased progressively in parallel with the increase in disease duration (Figure 1).
Table 1. Demographic features and SDI scores in 1,015 patients with juvenile-onset systemic lupus erythematosus*
|No. male/no. female||169/846|
|Age at diagnosis, years||11.9 ± 3.1 (12.2 [1.3–17.9])|
|Age at study visit, years||15.9 ± 4.1 (15.7 [2.4–34.8])|
|Disease duration at study visit, years||4.0 ± 3.6 (2.7 [0.5–26.7])|
|SDI score||0.8 ± 1.4 (0 [0–12])|
|No. (%) of patients with specific SDI score|| |
| >0||405 (39.9)|
| 1||213 (21.0)|
| 2||79 (7.8)|
| 3||50 (4.9)|
| 4||25 (2.4)|
| 5||19 (1.9)|
| 6||9 (0.9)|
| ≥7||10 (1.0)|
Figure 1. Mean and SD scores on the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) in patients with juvenile-onset systemic lupus erythematosus, by disease duration interval.
Download figure to PowerPoint
Table 2 shows the frequency of damage in the 12 organ systems/domains of the SDI along with the frequency of growth failure and delayed puberty. For each domain, the frequency in the whole patient population and the frequency in patients divided into 4 different disease duration intervals are provided. Among the SDI domains, renal damage, neuropsychiatric damage, and musculoskeletal damage were observed most frequently, followed by ocular damage and skin damage. Nephrotic-range proteinuria was the most frequently observed SDI item (10.1%), followed by cataracts (5.9%), cognitive impairment or major psychosis (5.6%), muscle atrophy or weakness (4.5%), scarring chronic alopecia (4%), avascular necrosis of bone (3.7%), and seizures (3.7%). Growth failure and delayed puberty were observed in 15.3% and 11.3% of the patients, respectively. The frequency of damage in most domains increased markedly with the increase in disease duration, particularly after the fifth year. However, damage in the neuropsychiatric, renal, and musculoskeletal organ systems occurred at a high rate in the first year of disease. The frequency of both growth failure and delayed puberty increased significantly with the increase in disease duration.
Table 2. Frequency of damage in the 12 organ systems/domains of the SDI and frequency of growth failure and delayed puberty in the study patients examined as a whole and divided into 4 disease duration intervals*
| ||All patients (n = 1,015)||Disease duration, years|
|0.5–1 (n = 172)||1–3 (n = 370)||3–5 (n = 181)||>5 (n = 292)||P|
|SDI domain|| || || || || || |
| Ocular||83 (8.2)||6 (3.5)||15 (4.1)||11 (6.1)||51 (17.5)||<0.001|
| Neuropsychiatric||109 (10.7)||15 (8.7)||32 (8.6)||16 (8.8)||46 (15.8)||0.013|
| Renal||132 (13.0)||14 (8.1)||34 (9.2)||27 (14.9)||57 (19.5)||<0.001|
| Pulmonary||18 (1.8)||2 (1.2)||5 (1.4)||3 (1.7)||8 (2.7)||0.51|
| Cardiovascular||30 (3.0)||1 (0.6)||6 (1.6)||5 (2.8)||18 (6.2)||0.001|
| Peripheral vascular||49 (4.8)||2 (1.2)||9 (2.4)||11 (6.1)||27 (9.2)||<0.001|
| Gastrointestinal||22 (2.2)||2 (1.2)||9 (2.4)||2 (1.1)||9 (3.1)||0.38|
| Musculoskeletal||109 (10.7)||11 (6.4)||36 (9.7)||14 (7.7)||48 (16.4)||0.002|
| Skin||77 (7.6)||5 (2.9)||20 (5.4)||15 (8.3)||37 (12.7)||<0.001|
| Premature gonadal failure||29 (2.9)||3 (1.7)||5 (1.4)||5 (2.8)||16 (5.5)||0.011|
| Diabetes (regardless of treatment)||4 (0.4)||0||1 (0.3)||1 (0.6)||2 (0.7)||0.66|
|Growth failure†||144 (15.3)||15 (8.7)||42 (11.6)||24 (15.2)||63 (25.2)||<0.001|
|Delayed puberty‡||63 (11.3)||7 (6.4)||25 (9.9)||12 (12.1)||19 (19.7)||0.02|
The SDI items that were detected in <5 patients (<0.5%) are listed in Table 3. These uncommon items included 2 of the 12 organ system/domains (diabetes and malignancy), 3 of the 5 pulmonary items (shrinking lung, pleural fibrosis, and pulmonary infarction or resection not for malignancy), 3 of the 5 cardiovascular items (angina or coronary artery bypass, myocardial infarction, and cardiomyopathy), 1 of the 4 peripheral vascular items (claudication for 6 months), 4 of the 5 gastrointestinal items (mesenteric insufficiency, chronic peritonitis, stricture or upper gastrointestinal surgery, and pancreatic insufficiency), and 1 of the 6 musculoskeletal items (ruptured tendons). In addition, a score of 2, which means involvement of >1 site, for significant tissue loss and for infarction or resection of bowel, spleen, liver, or gall bladder ever, was recorded in <0.5% of the patients.
Table 3. SDI domains and items that were observed in <5 (<0.5%) of the patients*
|Angina or coronary artery bypass||0|
|Pulmonary infarction or resection not for malignancy||1 (0.1)|
|Mesenteric insufficiency||1 (0.1)|
|Stricture or upper gastrointestinal surgery||1 (0.1)|
|Significant tissue loss, >1 site||2 (0.2)|
|Chronic peritonitis||2 (0.2)|
|Pleural fibrosis||3 (0.3)|
|Claudication for ≥6 months||3 (0.3)|
|Pancreatic insufficiency||3 (0.3)|
|Infarction or resection of bowel, spleen, liver, or gall bladder ever, >1 site||4 (0.4)|
|Shrinking lung||4 (0.4)|
Taking into account the specific aspects of SLE in children and adolescents, we devised a modified version of the SDI for use in patients with juvenile-onset SLE, which we propose to name the Pediatric SLICC/ACR Damage Index (Ped-SDI) (Table 4). To ensure harmonization with the original SDI, we have not changed any existing items but included growth failure and delayed puberty as additional organ systems/domains. Thus, the Ped-SDI is composed of 14 organ systems/domains and includes 43 items; the maximum score possible is 49. The glossary of terms for the items of the original SDI (10, 11) has been maintained, with the sole exception of the addition of an indication that in younger children proteinuria should be adjusted for height and weight. Specific definitions for the new items growth failure and delayed puberty have been provided. Like the original SDI, the Ped-SDI is intended to capture the presence of damage, defined as persistent changes in anatomy, physiology, pathology, or function, which may be the result of prior active disease, complications of therapy, or comorbid conditions, are not due to currently active disease, and have been present for at least 6 months. Damage is often irreversible and cumulative, and thus, damage scores are most frequently expected to increase or remain stable over time. However, because some forms of damage may improve or even resolve in pediatric patients, it is anticipated that in some cases scores may decline.
Table 4. The Pediatric Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index*
|Ocular (either eye, by clinical assessment)|| |
| Any cataract ever||0 or 1|
| Retinal change or optic atrophy||0 or 1|
| Cognitive impairment (e.g., memory deficit, difficulty with calculation, poor concentration, difficulty in spoken or written language, impaired performance level) or major psychosis||0 or 1|
| Seizures requiring therapy for 6 months||0 or 1|
| Cerebrovascular accident ever (score 2 if >1), resection not for malignancy||0, 1, or 2|
| Cranial or peripheral neuropathy (excluding optic)||0 or 1|
| Transverse myelitis||0 or 1|
| Estimated or measured glomerular filtration rate <50%||0 or 1|
| Proteinuria ≥3.5 gm/24 hours||0 or 1|
| or|| |
| End-stage renal disease (regardless of dialysis or transplantation)||3|
| Pulmonary hypertension (right ventricular prominence or loud P2)||0 or 1|
| Pulmonary fibrosis (by physical and radiographic examination)||0 or 1|
| Shrinking lung (by radiographic examination)||0 or 1|
| Pleural fibrosis (by radiographic examination)||0 or 1|
| Pulmonary infarction (by radiographic examination) or resection not for malignancy||0 or 1|
| Angina or coronary artery bypass||0 or 1|
| Myocardial infarction ever (score 2 if >1)||0, 1, or 2|
| Cardiomyopathy (ventricular dysfunction)||0 or 1|
| Valvular disease (diastolic murmur, or systolic murmur >3/6)||0 or 1|
| Pericarditis for ≥6 months or pericardiectomy||0 or 1|
|Peripheral vascular|| |
| Claudication for ≥6 months||0 or 1|
| Minor tissue loss ever (pulp space)||0 or 1|
| Significant tissue loss ever (e.g., loss of digit or limb, resection) (score 2 if >1)||0, 1, or 2|
| Venous thrombosis with swelling ulceration, or venous stasis||0 or 1|
| Infarction or resection of bowel (below duodenum), spleen, liver, or gall bladder ever (score 2 if >1)||0, 1, or 2|
| Mesenteric insufficiency||0 or 1|
| Chronic peritonitis||0 or 1|
| Stricture or upper gastrointestinal tract surgery ever||0 or 1|
| Pancreatic insufficiency requiring enzyme replacement or with pseudocyst||0 or 1|
| Muscle atrophy or weakness||0 or 1|
| Deforming or erosive arthritis (including reducible deformities, excluding avascular necrosis)||0 or 1|
| Osteoporosis with fracture or vertebral collapse (excluding avascular necrosis)||0 or 1|
| Avascular necrosis (score 2 if >1)||0, 1, or 2|
| Osteomyelitis||0 or 1|
| Ruptured tendons||0 or 1|
| Scarring chronic alopecia||0 or 1|
| Extensive scarring or panniculum other than scalp and pulp space||0 or 1|
| Skin ulceration (excluding thrombosis) for >6 months||0 or 1|
|Diabetes (regardless of treatment)||0 or 1|
|Malignancy (exclude dysplasia) (score 2 if >1 site)||0, 1, or 2|
|Premature gonadal failure||0 or 1|
|Growth failure||0 or 1|
|Delayed puberty||0 or 1|
- Top of page
- PATIENTS AND METHODS
We investigated the pattern of accumulation of damage in a very large cohort of patients with juvenile-onset SLE. Since patients in 39 countries were enrolled in the present study, our results are likely generalizable to other series of patients with juvenile-onset SLE in pediatric rheumatology centers worldwide. We found that damage was present in 40% of the patients after a mean disease duration of 4 years, with a mean SDI score of 0.8. The observed frequency and amount of damage were lower than those reported in previous studies of juvenile SLE, excluding the study by Ravelli et al (7), which was incorporated in the present analysis. In the other studies (6, 8, 9) the frequency of damage and the mean SDI scores ranged from 59% to 61% and from 1.3 to 2.1, respectively. This may reflect the relatively short disease duration of many of our patients, 53% of whom had been followed up for <3 years after diagnosis. Disease duration has been found to correlate strongly with damage accumulation both in juvenile (7, 9) and in adult-onset (24–26) SLE. Indeed, the frequency of damage and the mean SDI score rose to 58.2% and 1.4, respectively, when we restricted the analysis to patients with a disease duration of >5 years, who represented approximately one-third of the study population.
Among the SDI organ systems/domains, renal damage was observed most frequently, followed by neuropsychiatric, musculoskeletal, ocular, and skin damage. Cardiovascular damage, gastrointestinal damage, pulmonary damage, premature gonadal failure, and diabetes were recorded in ≤3% of the patients, and no malignancies were observed. When we divided patients according to disease duration, we found that the neuropsychiatric system was the most frequently involved system in the group with disease duration of <1 year, and that the renal system was the most frequently involved system in the group with disease duration of >5 years. The frequencies of ocular, cardiovascular, peripheral vascular, and skin damage were greater in patients with disease duration of >5 years than in those who had been followed up for <5 years. Neuropsychiatric, renal, and musculoskeletal damage were relatively common among patients with disease duration of <3 years. Taken together, these findings suggest that neuropsychiatric, renal, and musculoskeletal damage may occur early in a number of patients with juvenile-onset SLE, whereas ocular, cardiovascular, peripheral vascular, and skin damage are more likely to develop later.
The main purpose of our study was to obtain information that could be employed to devise a modified version of the SDI for use in patients with juvenile-onset SLE. We first examined the frequency of 2 sources of damage, growth failure and delayed puberty, that are of critical importance in children and adolescents with SLE and are not included in the current SDI. Growth failure is a common complication of juvenile-onset SLE that may have a serious psychological impact. Children and adolescents may experience growth retardation as a result of high-dose corticosteroid therapy, although growth may catch up significantly once steroid doses are lowered. The degree of permanent growth retardation is variable and has been insufficiently studied. A high frequency of short stature (38%), defined as height below the fifth percentile for age, in a series of patients with juvenile-onset SLE was reported in 1990 (27). In the clinical setting, it is essential to regularly assess the level of growth in children with SLE, for monitoring (and, whenever possible, reducing) the accumulation of disease- and treatment-related damage over time. In addition, assessment of final height is needed to verify whether growth failure has resulted in permanently short height.
Although delayed puberty, which is related to active disease and/or treatment, is a temporary phenomenon, it may have irreversible consequences because it may hamper some important physiologic events, such as growth spurt or bone mass accretion. These losses may not be regained once puberty develops and may lead to final short stature and premature osteoporosis (28, 29). Delayed puberty may also cause important emotional and social difficulties. For example, it may result in young adults with the illness being treated based on how they appear physically rather than according to their emotional or intellectual development or chronological age. This may hamper their mastery of the necessary social skills needed to negotiate the adult environment (30).
Obesity, another potential consequence of the disease and its treatment, may increase the risk of cardiovascular and endocrine morbidity, may affect patients' quality of life, and has been associated with poorer disease outcome (31). For these reasons, we explored the frequency of obesity, defined as a body mass index ≥30 kg/m2 (32), in the 550 patients on whom data on height and weight were available. We found that 5.6% of these patients were obese (data not shown). Because obesity is not unique to children, is frequently observed in the general population, and the number of patients in whom it was disease related could not be established, we did not include it in the proposed pediatric SDI. However, we believe that this complication should be considered for inclusion in a future revision of the SDI.
The second step in the process was the identification of the SDI domains and items that are rarely seen in children and adolescents with SLE and may, therefore, be considered noncontributory to a pediatric damage index. We found that 2 domains (diabetes and malignancy) and a number of items in the pulmonary, cardiovascular, peripheral vascular, gastrointestinal, and musculoskeletal systems were present in <0.5% of patients. Furthermore, the item significant tissue loss and the item infarction or resection of bowel, spleen, liver, or gall bladder ever involved >1 site (yielding a score of 2) in <0.5% of patients. However, to ensure harmonization with the original SDI, particularly regarding the longitudinal assessment of damage after the transition of pediatric patients to adulthood, we have decided to provisionally maintain these items in the pediatric SDI. Future studies in additional patient series will help to establish whether they may be removed from the instrument without affecting its clinical reliability.
Based on these results and considering the distinctive aspects of SLE in children and adolescents, we devised a pediatric version of the SDI, which is shown in Table 4. The Ped-SDI includes 14 organ systems/domains and 43 items. As in the original instrument, damage is ascertained clinically and scored if it has occurred since disease onset and has been present for at least 6 months. However, unlike the original SDI, which defines damage as an irreversible change in an organ or system, the Ped-SDI definition takes into account the fact that some forms of damage are potentially reversible in pediatric patients, owing to their ability to recover and regenerate to a greater degree than adults. For instance, once better disease control is achieved, avascular necrosis of bone can improve due to restoration of the normal growth process. Similar considerations apply to the newly added domains growth failure and delayed puberty. Notably, it has been recognized recently that some of the damage items included in the SDI, such as premature gonadal failure (with resultant secondary amenorrhea), are occasionally seen to reverse clinically with time in adult patients with SLE (14).
We have retained the definitions for all the original SDI items. Although a redefinition of the cognitive impairment item has been advised to increase its applicability in younger patients (7), we believe it is suited as it stands as a screening item and that a more precise and age-specific assessment of the extent of neurocognitive impairment would require the use of specific neuropsychological testing, such as IQ or the Wechsler scale (33). As recently suggested (14), we have recommended that proteinuria be adjusted for height and weight in younger children.
Our study should be viewed in light of some potential limitations, which include its cross-sectional design. Better insights into the development of long-term damage can be obtained by studying cohorts of newly diagnosed patients observed prospectively over time. Because patients were assessed in many different centers worldwide, we cannot exclude the possibility of some lack of consistency in the collection of data. The evaluation of the development of secondary sexual characteristics was made in most centers by a pediatric rheumatologist and not by an endocrinologist, which may affect its reliability. Growth retardation was assessed using international standards; it is unclear whether these standards are applicable to children living in all countries. For this reason, our findings may not reflect the true prevalence of growth failure in patients with juvenile-onset SLE. The list of damage items is not intended to be exhaustive, but may be modified or enlarged after the application of the index to other populations of patients seen in different clinical or research settings. We anticipate that the Ped-SDI may undergo a process of refinement as we and others incorporate new data, including information on the changes in score over time.
In conclusion, we propose a modified version of the SDI for use in children and adolescents with SLE, which was developed through evidence-based scrutiny of potential sources of damage in 1,015 patients. This new instrument needs prospective validation in other populations of patients seen in different clinical or research settings.