Characteristics of all studies evaluating relationship between SLE disease activity and damage measures on the outcomes of mortality and damage accrual.
| Study (Composite Measure) | N | Design | Population | Outcome | Study Details | Newcastle-Ottawa Quality Rating (no. stars) |
|---|---|---|---|---|---|---|
| BILAG (British Isles Lupus Assessment Group score for disease activity; BILAG or BILAG 2004) | ||||||
| Furie, 20098 | 321 | RCT | Adult, phase II RCT (retrospective application of the SRI). | Non-renal morbidity | Belimumab treatment resulted in a statistically larger percentage of responders than treatment with placebo | Low risk of bias* |
| Hay, 19939 | 353 | Prospective observational cohort | Adult, multicentered, England. Median disease duration 9.5 yrs (range 0–58). | Non-renal morbidity | Validation of the BILAG Index (interrater, criterion, construct validity) | 6 |
| Lopez, 201210 | 350 | Prospective observational cohort | Adult, SLE clinic, London, UK. Median disease duration (at T0) 6 yrs (0–34). | Mortality, renal morbidity, damage | Mean total BILAG score associated with mortality, new organ damage and CV/pulmonary or musculoskeletal damage. Pre-existing SDI = independent predictor of mortality and further organ damage. | 8 |
| Stoll, 200011 | 141 | Prospective observational cohort | Adult, SLE outpatient clinic. Mean disease duration 10.2 yrs (SD 6.3). | Damage | Total damage score (OR 1.46, 95% CI 1.04–2.05) and health perception (OR 0.96, 95% CI 0.93–0.99) at inclusion predicted damage at 3 yrs | 6 |
| Stoll, 200412 | 141 | Prospective observational cohort | Adult, SLE clinic. Mean disease duration 10.2 yrs (SD 6.3). Same clinic as reference 10. | Mortality, damage | Multiple logistic regression analysis showed that high total disease activity over entire study period predicted death and increase in damage (p < 0.001); replacement of total BILAG score by average number of A-flares predicted accrual of damage during study period (p = 0.004) | 6 |
| SLEDAI (SLE Disease Activity Index), SELENA-SLEDAI, SLEDAI-2K, Mexican SLEDAI, Adjusted Mean SLEDAI (AMS), Weight-adjusted SLEDAI (WAS) | ||||||
| Bandeira, 200613 | 57 | Retrospective observational cohort | Pediatric, consecutive patients 1988–2000 in 3 sites (2 Italy, 1 Brazil). Followed for ≥ 3 yrs within 12 mos of diagnosis. | Damage (SDI) | Patients accruing new damage over 3 yrs had greater frequency of severe disease flare. Damage accrual associated with severe disease flares. | 8 |
| Becker-Merok, 200614 | 86 | Prospective observational cohort | Adult, Tromso (Norwegian) Lupus Cohort. Mean disease duration 11.9 yrs (0.5–38). | Mortality, damage (SDI) | SDI scores higher in 37 nonsurvivors (23.4%; SDI 2.1) vs survivors (SDI 0.9; p < 0.05); damage accrual linear in first decade of disease; only independent predictor for SDI ≥ 3 was WAS score > 3 (HR 2.34, 95% CI 1.1–4.9); age > 40 yrs at diagnosis (HR 5.6, 95% CI 2.4–12.7) and WAS > 3 (HR 2.4, 95% CI 1.2–4.0) = significant predictors of death | 8 |
| Bruce, 201515 | 1722 | Prospective observational inception cohort | Adult, SLICC Cohort Study. Enrolled within 15 mos of diagnosis. | Damage, mortality | SDI ≥ 1 at enrollments associated with worsening of SDI (vs those with SDI 0; p < 0.01); SLEDAI-2K score associated with damage development and progression. SDI associated with HR 1.46 (95% CI 1.18–1.81) for mortality. | 8 |
| Clowse, 201316 | 1478 | Multiple clinical trials in SLE | LCTC Lupus Data Registry. Mean disease duration: SLE with damage accrual = 16.8 yrs vs stable damage = 12.17 yrs. | Damage | Baseline SLEDAI ≥ 10 associated with increase in SDI (RR 3.66, 95% CI 1.96–6.84). MVA showed only baseline SDI and SLEDAI predict damage accrual | 8 |
| Feng, 201117 | 1956 | Retrospective observational cohort | Adult, medical records of hospitalized SLE patients in China (15 hospitals). Median age disease onset: 30 yrs. | Mortality | SLEDAI > 8 at discharge independent predictor of mortality (HR 1.64, 95% CI 1.12–2.42; p = 0.012) | 7 |
| Gilboe, 200118 | 93 | Prospective observational cohort | Adult, Norwegian hospital SLE patients. Baseline 1995 and followup 2 yrs later; mean disease duration 6.1 (0–31) yrs. | Damage | Baseline SLEDAI OR 1.14 (1.00–1.28, p = 0.04) and SDI OR 1.52 (1.02–2.27, p = 0.04) for change in organ damage | 8 |
| Hill, 201119 | 1168 | Prospective observational cohort | Adult, Hopkins cohort. Disease duration 6 yrs; 12-mo observation period. | Damage (SDI), death | AMS predicted death: HR 1.23 (95% CI 1.14–1.33), new onset renal damage: HR 1.24 (95% CI 1.08–1.42), new onset CV disease: HR 1.18 (95% CI 1.08–1.30); not predictor of new onset overall damage (HR 1.05, 95% CI 1.00–1.11) | 7 |
| Ibanez, 2005/2006/200720,21,22,† | 575 | Prospective observational cohort | Adult, Toronto Lupus Cohort. Disease duration last clinic visit 10.6 ≥ 8.1 yrs, 1970–2002. | Mortality, damage | AMS (HR 1.16, p < 0.0001) and age at diagnosis (HR 1.05, p < 0.0001) predicted survival; AMS (HR 1.06, p < 0.0001), age at diagnosis (HR 1.02, p < 0.0004), and disease duration (HR 1.05, p < 0.0001) predicted damage | 8 |
| Liang, 201023 | 43 | Retrospective observational cohort, patients receiving PD | Adult, 6 patients < 20 yrs old, Taiwan. Duration of PD 39.7 ± 22.4 mos. | Mortality | Pre-dialysis non-renal SDI and non-renal SLEDAI ≠ predict mortality in univariate Cox regression analysis. No difference in SDI scores between the dead and living groups. | 6 |
| Lilleby, 200524 | 71 | Retrospective observational cohort, and cross-sectional | Juvenile. Mean disease duration 10.8 ± 8.2 yrs. | Damage, mortality | 4 died with higher last SDI score vs those who lived (3.3 vs 1.3, p = 0.012); increasing SDI not associated with SLEDAI in MVA | 6‡ |
| Lin, 201225 | 158 | Retrospective observational cohort | Adult, West China Hospital. Mean disease duration 63.86 ± 48.17 mos. | Mortality, damage | SLEDAI at diagnosis associated with survival in late-onset SLE (50 yrs+); OR 1.091 (1.030–1.155, p = 0.003) | 7 |
| Mikdashi, 200426 | 130 | Prospective observational cohort | Adult, University of Maryland Lupus Cohort. Mean disease duration: NP 0 = 7.2 (2.6–11.8) yrs vs NP > 1 = 8.5 (3.4–13.6) yrs | NP damage | SLEDAI independent predictor of significant NP damage | 7 |
| Mok, 200327 | 242 | Prospective observational cohort | Adult, Hospital, Hong Kong. Mean disease duration 75.3 ± 79 mos; 3 yrs followup. | Mortality, damage | Increase in SDI scores from baseline predictive of mortality OR 1.47 per 1 point, 95% CI 1.03–2.11, p = 0.04. No. major disease flares and current/past CYC treatment independently predicted damage accrual over 3 yrs. | 7 |
| Mok, 200628 | 282 | Prospective observational cohort | Adult, possible pediatric, hospital, Hong Kong. Mean followup 6.7 yrs. | Mortality, NP damage (SDI) | SLEDAI at diagnosis, cum non-NP damage, + antiphospholipid antibodies, ever use of methylprednisolone independently predicted NP damage in MVA; logistic regression did not confirm association between early or cumulative NP damage and mortality | 8 |
| Nossent, 199329 | 68 | Prospective observational cohort | Adult, possible pediatric; in/outpatients, hospital, Caribbean. Mean followup 38 (± 33) mos. | Mortality, disease exacerbations | High persistent disease activity (weighted average of SLEDAI scores > 10) independently associated with decreased survival | 6 |
| Nossent, 201030 | 200 | Prospective inception cohort | Adult, possible pediatric, multicenter European study. New-onset SLE. Followed for 5 yrs. | Disease damage (SDI) | MVA showed that persistent disease activity (average annual SLEDAI) predicted damage accrual (SDI ≥ 1) | 8 |
| Petri, 201231 | 2054 | Prospective observational cohort | Adult, Hopkins cohort. Mean age at diagnosis 33 yrs; mean followup 6.4 yrs. | Damage (SDI) | SDI increased by 0.13 per year; higher disease activity associated with more damage during followup; however, effect of disease activity lost in MVA | 7 |
| Pons-Estel, 2004 (SLEDAI/Mexican SLEDAI)32 | 1214 | Prospective, observational cohort | Adult/pediatric, GLADEL cohort. Median disease duration 32 mos. | Mortality | Max mean SLEDAI/Mexican SLEDAI and SDI were significantly higher in dead (34 pts) vs living (1180 pts); mortality predicted in stepwise logistic regression model including SDI (≥ 1 vs 0; OR 2.8, 1.2–6.4) | 8 |
| Ramirez Gomez 2008 (SLEDAI/Mexican SLEDAI)33 | 1214 | Prospective observational | Pediatric and adult GLADEL cohort. Mean followup: pediatric 1.7 (0.8–2.9) yrs; adults 1.6 (0.8–2.7) yrs. | Mortality | Death in first 5 yrs: disease activity and infection. Children had higher disease activity scores (p = 0.001); adults had greater disease damage (p = 0.02). | 8 |
| Suarez-Larios, 2013 (Mexican SLEDAI)34 | 56 | Retrospective observational cohort | Pediatric, ICU medical records, Mexico City (Jan 1999–Dec 2008). | Mortality | Mortality associated with high SLEDAI score; main cause of death = infection | 7 |
| Telles, 201335 | 179 | Prospective observational cohort | Adult, Brazil outpatient clinic (2006). Median disease duration at T0 8.2 (4.3–12.4) yrs. | Mortality | Higher modified SLEDAI-2K (HR 1.12, 95% CI 1.01–1.25; p = 0.040) and SDI (HR 1.57, 95% CI 1.23–2.01; p < 0.001) among dead than survivors; initial SDI ≥ 3 significantly increased risk of death (log rank: p < 0.001) | 8 |
| Uziel, 200736 | 102 | Retrospective observational cohort | Pediatric, Israeli Pediatric Rheumatology Internet Registry (1987–2003). | Damage | Initial SLEDAI predicted development of late damage, while no other factors were predictive; 51% had minimum 5 yrs followup | 8 |
| Wu, 201437 | 665 | Prospective observational cohort | Adult, Chinese hospital SLE admissions (Jan 2006–Dec 2009). | Mortality | MVA with confounder adjustments found that male sex, older age at onset, high SLEDAI scores at time of diagnosis were independent risk factors for all-cause mortality | 8 |
| Wu, 201438 | 104 | Prospective observational cohort | Pediatric, tertiary care center, Taiwan, fulfilling ACR criteria and had renal biopsy (Jan 1999–Dec 2011). | Worsening renal function | PLN patients with poorer outcomes; prognostic factors = high baseline SLEDAI-2K (> 20; HR 6.76, p = 0.002), baseline GFR < 60 ml/min/m2 (HR 3.88, p = 0.022); early responder (HR 0.19, p = 0.013) | 8 |
| Zonana-Nacah, 200739 | 41 | Observational cohort (not specified) | Hospitalized SLE patients in Mexico between 2004 and 2006. Mean followup 9.7 ± 6 mos. | Mortality | 16 (39%) died; they had significantly higher SLEDAI (p = 0.004) and SLICC (p = 0.03) scores | 8 |
| SLAM (SLE Activity Measure) | ||||||
| Alarcon, 200140 | 258 | Prospective observational cohort | Adult, met ACR criteria, 5 years or less. | Mortality | SLAM at enrollment OR 1.09 (1.01–1.17, p = 0.0194) and SDI (first computed) OR 1.45 (1.19–1.91, p = 0.0094) predictors of mortality in MVA | 6 |
| Karlson, 199741 | 200 | Retrospective observational cohort | Adult, 5-center, multiple insurance. Disease duration < 7 yrs. | Organ damage (SDI), SF-36 | MVA showed that greater SLAM at diagnosis associated with greater damage; greater damage at diagnosis associated with greater damage | 5 |
| Nieves-Plaza, 201142 | 61 | Retrospective observational cohort | Adult, University of Puerto Rico. | Renal disease progression | SLAM-R and SDI scores did not predict a decline in GFR. SLAM score > 8 related to renal deterioration (HR 1.55, 95% CI 0.39–3.6). | 5 |
| Peschken, 200943 | 1416 | Prospective observational cohort | Adult, pediatric, Canadian (1000 Faces). Mean disease duration 12 ± 10 yrs. | Damage | MVA confirmed association of SLAM-R as independent predictor of damage accumulation in addition to low income, disease duration, age, CYC and prednisone ever | 7 |
| Toloza, 200444 | 158 | Prospective observational cohort | Adult, LUMINA median followup 24 (5–112) mos. | Time to initial damage | Higher SLAM scores were independent predictors of shorter time to initial damage in MVA (HR 1.09, 95% CI 1.04–1.15) | 8 |
| ECLAM (European Consensus Lupus Activity Measurement Index) | ||||||
| Shariati-Sarabi, 201345 | 71 | Inception cohort, first renal symptoms | Adult, Iranian Medical Center (2005–2011). Mean disease duration 1 mo–15 yrs. | Damage (SDI) | SLEDAI-2K significantly associated with ECLAM results in correlation analyses (r = 0.827, p < 0.001); SDI significantly related to SLEDAI-2K and ECLAM (r = 0.699, p < 0.001) | 5 |
| Systemic Lupus International Collaborating Clinics (SLICC) ACR Damage Index (SDI) | ||||||
| Alarcon, 200446 | 352 | Prospective observational cohort | Adult, LUMINA cohort. Mean disease duration 18.3 (16.2) mos. | Damage | SLAM score (p < 0.0001) and prior SDI (p < 0.0001) associated with damage accrual in MVA | 8 |
| Appenzeller, 200547 | 55 | Retrospective observational cohort | Pediatric SLE cohort Brazil. Mean disease duration 4.8 (4.7) yrs. | Mortality | Male sex, infection, and nephritis independently associated with death in MVA; SDI did not influence survival | 8 |
| Bruce, 201515 | 1722 | Prospective observational inception cohort | Adult, SLICC Cohort Study. | Damage, mortality | See SLEDAI section | 8 |
| Cardoso, 200848 | 105 | Prospective observational cohort | Adult, consecutive patients meeting ACR class criteria. Median followup 6.3 yrs. | Mortality | Initial and final SDI ≥ 3 = independent predictors of mortality; HR 3.0 (95% CI 1.1–8.2) and 4.7 (95% CI 1.6–14.5); damage accrual during followup strongest predictor of death (HR 5.1, 95% CI 2.0–13.0) | 8 |
| Chambers, 200949 | 232 | Retrospective observational cohort | Adult, outpatient SLE clinic. Minimum 10 yrs followup. | Mortality | Increase in damage score associated with higher risk of death overall; for every 1-point increase in damage score, the patient was 1.32-times more likely to die; HR 1.32 (95% CI 1.09, 1.60; p < 0.005). Adjusted HR (adjusted for age at SLE onset) is 1.40 (1.14, 1.72). | 8 |
| Clowse, 201316 | 1478 | Multiple clinical trials | LCTC Lupus Data Registry. | Damage | See SLEDAI section | 8 |
| Danila, 200950 | 635 | Prospective observational cohort | Adult, LUMINA cohort. Mean disease duration 5.7 (3.7) yrs. | Mortality | Excluding poverty from MVA, renal domain of SDI was independently associated with shorter time to death (HR 1.65; 95% CI 1.03–2.66) | 8 |
| Gilboe, 200118 | 93 | Prospective observational cohort | Adult, Norwegian hospital SLE patients. | Damage | Study details in SLEDAI section | 8 |
| Gladman, 20005 | 1297 | Prospective observational cohort | Adult, SLICC patients. Mean age at diagnosis 32 (3–93) yrs. | Mortality | Patients who died had higher SDI scores early in their course (1.56) vs patients who remained alive (0.99; p = 0.0003) | 8 |
| Liang, 201023 | 43 | Retrospective observational cohort, PD patients | Adult, 6 patients < 20 yrs old. | Mortality | Study details in SLEDAI section | 6 |
| Lilleby, 200524 | 71 | Retrospective observational cohort, cross-sectional | Pediatric, Norwegian hospital. | Mortality | Study details in SLEDAI section | 7‡ |
| Lopez, 201210 | 350 | Prospective observational cohort | Adult, SLE clinic, London, UK. | Mortality, renal morbidity, damage | Study details in SLEDAI section | 8 |
| Mak, 200752 | 149 | Retrospective observational cohort | Adult/pediatric (1991–2003). | Renal damage (renal SDI) | Late-onset (≥ 50 yrs) SLE patients accrued more renal damage but not significantly associated with age after MVA | 6 |
| Manger, 200253 | 338 | Prospective observational cohort | Adult, University of Erlangen-Nuremberg. Median disease duration 7.8 yrs. | Mortality | Increase of ≥ 2 points of SDI from first to third yr of disease associated with mortality (RR 7.7, 3.3–18.6, p < 0.0001) | 8 |
| Mok 2013, 2014 (2 studies) 54,55,56† | 694, 747, 756 | Prospective observational cohort | Adult/pediatric, outpatient, Hong Kong (1995–2011); mean followup since diagnosis 9.6 ± 7.3 yrs. | Mortality, damage (SDI) | Age- and sex-adjusted HR of mortality in SLE patients are 2.23 (1.29, 3.85) with renal disease, 3.59 (2.20, 5.87) with renal damage, and 9.20 (4.92,17.2) with endstage renal disease; Cox regression revealed that early damage associated with adjusted HR 6.49 (95% CI 3.84–11, p < 0.001) | 8 |
| Nived, 200257 | 80 | Prospective observational cohort | Adult SLE patients in Orup, Sweden (diagnosed 1981–91). | Mortality | 5-yr SDI of ≥ 2 increased mortality risk by 3.4 (1.5–14.4) | 8 |
| Pons-Estel, 2004, (Mexican SLEDAI)32 | 1214 | Prospective observational cohort | Adults/pediatric GLADEL cohort. | Mortality | See SLEDAI section | 8 |
| Rabbani, 201058 | 198 | Prospective observational cohort | Adult/pediatric, Aga Khan University Hospital, followed for 10 yrs. | Mortality | 1-yr postdiagnosis mean renal damage score was significant predictor of death within 10 yrs of diagnosis | 8 |
| Rahman, 200159 | 263 | Prospective observational cohort | Adult, University of Toronto Lupus Clinic. Within 1 year of diagnosis prior to 1988; 10-year followup or death. | Mortality | Renal damage (p = 0.013) and trend to CV disease (p = 0.056) seen more often in SLE patients who died within 10 yrs (p = 0.013) than living patients; early damage associated with greater mortality after 10 yrs of followup | 8 |
| Stoll, 199660 | 80 | Retrospective, inceptional cohort | Adult, Switzerland. Within 10 yrs of diagnosis. | Mortality, damage | 1 yr post-diagnosis mean renal damage score predicted endstage renal failure and mean pulmonary damage score significantly predicted death within 10 yrs of diagnosis | 6 |
| Telles, 201335 | 179 | Prospective observational cohort | Adult, Brazil SLE outpatient clinic. | Mortality | Study details in SLEDAI section | 8 |
| Zonana-Nacah, 200739 | 41 | Observational cohort (not specified) | Hospitalized SLE patients in Mexico. | Mortality | Study details in SLEDAI section | 8 |
| BILD (Brief Index of Lupus Damage) | ||||||
| Katz, 201461 | 958 | Observational cohort (not specified) | Adult, UCSF Lupus Outcomes Study. Mean disease duration 16 ± 9 yrs. | Mortality | Higher risk of death with BILD scores of 2 (HR 6.1, 95% CI 1.3–30.0) and ≥ 3 (HR 10.8, 95% CI 2.5–46.2) | 8 |
↵* Cochrane Risk of Bias tool used to assess quality of RCT62.
↵† Three publications from same cohort, reported once.
↵‡ Adjusted version of Newcastle-Ottawa Scale for cross-sectional studies. SLE: systemic lupus erythematosus; RCT: randomized controlled trial; SRI: SLE Responder Index; CV: cardiovascular; LCTC: Lupus Clinical Trials Consortium Inc.; MVA: multivariate analysis; PD: peritoneal dialysis; NP: neuropsychiatric; cum: cumulative; CYC: cyclophosphamide; PLN: pediatric lupus nephritis; GLADEL: Grupo Latinoamericano de Estudio del Lupus; ICU: intensive care unit; ACR: American College of Rheumatology; GFR: glomerular filtration rate; SF-36: Medical Outcomes Study Short Form-36 questionnaire; LUMINA: Lupus in Minority: NAture vs Nurture cohort; UCSF: University of California at San Francisco; SELENA: Safety of Estrogens in Lupus Erythematosus National Assessment; SLAM-R: SLAM-Revised.