Background/Purpose:

Lupus nephritis (LN) is an important cause of morbidity and mortality in patients with systemic lupus erythematosus (SLE). An initial renal biopsy (RB) is currently recommended by most experts, but clinical practice varies. With recent changes in treatment paradigms for LN the role of RB in deciding treatment has been questioned. In addition, a recent systematic review found the interpretation of RB findings to be unreliable. Therefore, we systematically reviewed the evidence for an initial biopsy in LN patients.

Methods:

Medline, Embase, Scopus, and Cochrane Library databases were searched using the search terms systemic lupus erythematosus, lupus nephritis and renal biopsy among others. The search was English-language restricted; randomized controlled trials (RCTs), observational, case control, cohort and cross-sectional studies were included. The quality of the studies was assessed using the Newcastle-Ottawa Scale and Quality in Prognostic Studies tool. Two reviewers reviewed citations for inclusion/exclusion at each stage (title, abstract and full text review).

Results:

A total of 776 references were identified after removal of duplicates. No relevant RCTs were found. No studies directly examined the role of RB compared to clinical parameters (CP) in deciding treatment. No studies assessing renal outcome were found that included a comparator arm of patients without RB. Two related studies showed that RB added marginally to short and long-term prediction of renal outcome when added to CP alone; however these were >30 years old and used WHO criteria resulting in high risk of bias. Fifteen studies looked at clinicopathologic correlation of RB findings and CP. Most studies were relatively small, with large variation in populations studied, CP reported, and outcome measures described; 7 were >25 years old. Results were conflicting. Several studies found that poor CP (acute renal failure, hypertension) showed good correlation with Class IV & V histology on RB, as well as poor renal survival. No clear correlation between degree of proteinuria and LN biopsy class was found. Three studies found that in the setting of low grade proteinuria and normal renal function, many patients had Class III, IV, or V LN on RB; 2 studies did not find a correlation between RB findings and proteinuria. Three studies found higher titres of anti-dsDNA antibodies correlated with proliferative LN, 1 study in Hispanic patients did not. Two studies found low C4 and higher c1q antibodies correlated with proliferative LN on RB. Two studies found that RB findings predicted renal outcome, 2 did not. In 4 studies, RB identified non-LN causes of renal disease; ranging in frequency from 5-9%. Four citations found thrombotic microangiopathy in addition to LN in 30-50% of patients.

Conclusion:

The evidence that an initial RB changes management or prognosis in LN is of low quality due to high risk of bias. There is some evidence that RB may serve to identify non-SLE glomerular or other changes that influence management and outcome. As RB remains the clinical standard, is generally required for entry into clinical trials, and is being considered for inclusion in classification criteria, further research into the role, utility and reliability of RB is required.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/indications-for-initial-renal-biopsy-in-systemic-lupus-erythematosus-a-systematic-review-of-the-literature/