Article Text

Download PDFPDF
A7.9 Repeat renal biopsies help to tailor immunosuppression in lupus nephritis
  1. A Pakozdi1,
  2. R Rajakariar2,
  3. M Sheaff3,
  4. D Pyne1
  1. 1Department of Rheumatology, Barts Health NHS Trust, London, UK
  2. 2Department of Renal Medicine, Barts Health NHS Trust, London, UK
  3. 3Department of Histopathology, Barts Health NHS Trust, London, UK


Background and objectives Lupus nephritis (LN) is the major cause of morbidity and mortality in patients with systemic lupus erythematosus. The role of repeat kidney biopsies (RB) to guide treatment or to predict outcome and prognosis has been controversial. In this retrospective study we focused on histological characteristics of RBs and aimed to identify any clinical variables useful to predict histological changes.

Methods In a large single-centre cohort of 257 patients from 1988–2014 with biopsy proven LN, 58 (23%) had two or more biopsies (a total of 68 RBs). LN classes based on glomerular pathology were defined according to the ISN/RPS classification. Clinical and laboratory data were obtained from electronic records of patients.

Results The median time between initial and RB was 33 months [IQR, 15–84]. Caucasians (n = 7) had a lower RB rate of 14% compared to blacks (n = 36, 32%; p = 0.010). Indication for RB was worsening proteinuria (n = 38, 71%; of which 23 had associated rising creatinine, 61%), rise in serum creatinine alone (n = 6, 11%) and lack of treatment response (n = 9, 17%) defined as <50% reduction in proteinuria. At time of RB, 25 (78%) had raised dsDNA, 33 (73%) had low complements. LN class transition occurred in 33 (49%), most commonly from class II or V to III or IV (n = 11, 36%). Swith from a proliferative to a pure non-proliferative class was rare (n = 3, 4.4%). 5 RB (7%) showed inactive lesions either due to FSGS or advanced sclerosing LN. 42 (65%) had a change in their treatment regime. Immunosuppression was more likely to be escalated in case of a class switch (93%, p = 0.000). The histological transition could not be predicted by any serological or biochemical variables.

Conclusion Over a 1/3 of our LN patients showed histological transition to a more aggressive class, based on which the majority (93%) had treatment escalation. Histological transition could not be predicted by clinical values. Hence, we conclude that RB remains an important tool to guide management of selected patients with LN, in particular those with initial class II or V who flare.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.