A Retrospective Renal Study from a Lupus Vasculitis Clinic

Aim: In July 2009, a combined Renal Rheumatology Lupus Vasculitis (RRLV) clinic, the first of its kind for adult patients in Australia, was started at Royal Brisbane & Women’s Hospital. This is an audit of progression of renal disease to assess if patients attending this clinic had comparable results to published studies of similar cohorts with lupus Nephritis (LN) and vasculitis. Methods: We conducted a retrospective audit of all the patients who attended this clinic from July 2009 to October 2013. There were 33 patients followed up in the vasculitis group and 36 in the LN group. Patients with other connective tissue disorders were excluded from the analysis as the numbers were insignificant. Results: The mean estimated glomerular filtration rate of vasculitis and LN patients improved from 32.06 to 45.82 ml/min/1.73m2 and 62.42 to 65.53 ml/min/ 1.73m2 respectively. The mean urine protein/creatinine ratio of vasculitis and LN patients improved from 420 to 85 and 406 to 70 respectively. No patients died in either group. One vasculitis and two LN patients required maintenance dialysis. Three LN patients underwent renal transplantation. Conclusion: The results show excellent patient and renal survival and support the concept of a combined renal rheumatology clinic in managing renal disease from systemic connective tissue disorders.

Patients with CTDs have multi-system involvement and need care from Rheumatology and Renal Medicine.There are a few combined Lupus Vasculitis clinics in UK, USA, and for children, in Australia [6]- [8].A Renal Rheumatology Lupus Vasculitis Clinic (RRLV) was started at Royal Brisbane and Women's Hospital (RBWH) in July 2009.
The patients in this clinic were managed as per international guidelines (Tables 1-3) and detailed records on patient progress were maintained [9]- [18].Patient survival and progression of kidney disease were measured as key performance indicators.
We conducted a retrospective audit of progression of renal disease in patients attending the combined RRLV clinic to establish whether patients followed up in this clinic had comparable patient and renal survival rates with published reports of cohorts with lupus Nephritis (LN) and vasculitis.

Methods
This is an audit of all patients who attended this clinic from July 2009 to October 2013.They were grouped based on renal involvement with vasculitis or LN and analyzed separately as the reported survival rates are different [2] [3].Other CTDs were excluded from the analysis as the numbers were small.We studied trends in estimated glomerular filtration rate (eGFR), urine protein creatinine ratio (uPCR), patient and renal survival.

Data
Data collected from medical records included age, gender, cause of renal involvement, renal biopsy, pulmonary involvement and requirement of plasmapheresis and dialysis, were de-identified and recorded on a spread sheet.
Laboratory test results were recorded at six monthly intervals, along with data regarding mortality, renal loss and renal replacement modalities.Duration and details of follow up at renal-rheumatology clinic were collected which included date of first visit, number of clinic visits and total duration of follow up.

Statistical Methods
Descriptive statistics and frequency distributions were done for continuous and categorical variables respectively.Associations between eGFR and risk factors were calculated by bi-variate (unadjusted) and multivariate analysis (adjusted).A generalized estimating equations (GEE) model was used to analyze repeated measures on eGFR for the same patient with population-averaged effects of covariates.Exchangeable correlation structure was used.The data were analyzed by using SPSS 16.0 for Windows.

Patient Characteristics
There were 31 patients followed up with renal vasculitis and 36 with LN.Fifteen patients with miscellaneous CTD were excluded from the analysis due to small numbers.Among these, there were sixpatients with overlap syndrome, three patients each with Sjogren's syndrome, two patients with Henoch-Schonlein purpura, and one each with scleroderma, polyarteritis nodosa, cryoglobulinemic vasculitis and primary antiphospholipid antibody syndrome.
Among vasculitis patients, the number of patients with microscopic polyangiitis (MPA), granulomatous polyan-giitis (GPA), anti-glomerular basement membrane (anti-GBM) disease and ANCA negative disease were 12, 8, 2 and 3 respectively.Six patients did not undergo a biopsy and there were no patients with combined ANCA and anti-GBM antibodies.Among LN patients, number of patients with class I, II, III, IV and V disease on histology were 1, 3, 6, 16 and 3 respectively.Five patients did not have a biopsy and two had mixed class III and V disease on histology.
The patient characteristics namely age, gender, comorbidities, organ system involvement, eGFR and uPCR at baseline and at the end of follow up, duration of disease and clinic follow up are shown in Table 4.

Results
We analyzed data from patients with renal vasculitis and LN separately.During follow up in the vasculitis patients the mean eGFR improved from 32.06 to 45.82 ml/min/1.73m 2 (Figure 1(a)).The mean uPCR declined from 420 to 85 (Figure 1(b)).There were no deaths while one patient required maintenance dialysis.We conducted a GEE analysis of the trend of eGFR in patients with vasculitis for age, gender, renal histology, and plasmapheresis (Table 5).We found a statistically significant increase in mean eGFR in males as compared to females while with ages less than or greater than 60, there was no difference.During follow up, compared to patients MPA, patients with GPA and anti-GBM disease had a statistically significant increase in mean eGFR and patients with ANCA negative vasculitis did not show a significant change.Among patients requiring plasmapheresis the mean improvement in eGFR was significantly less as compared to patents who did not receive plasmapheresis.
In the LN group the mean eGFR improved from 62.42 to 65.53 ml/min/1.73m 2 (Figure 1(c)) and the mean uPCR from 406 to 70 (Figure 1(d)).There were no deaths, but 5 patients lost kidney function with 3 receiving renal transplantation and 2, maintenance dialysis.
A GEE analysis was conducted with regards to variables including, age gender, and histology.In LN group there was no statistically significant difference in the trend in eGFR based on age less than or greater than 60 or sex (Table 5).As compared to class V LN, class I LN had a mean improvement of 19.80 ml/min/1.73m 2 (P value 0.072) during follow up, not reaching statistical significance.Similarly there was no statistically significant difference in mean improvement in eGFR among other classes of LN.

Discussion
We compared results of our study with published data on patient survival and surrogate markers including renal  survival, trend in eGFR and uPCR for vasculitis and LN patients respectively.In a study of 273 ANCA vasculitis patients, survival rates at 1, 5 and 10 years were 90%, 83% and 74% respectively, similar to results from previous studies [2] [19].At presentation 48% required dialysis, and among those who were independent of dialysis 7% developed ESRD [2].In our clinic during 4 years of follow up, one patient with vasculitis required maintenance dialysis while none died.
In a study involving 491 patients with LN, the overall cumulative probability of survival at 1, 5, 10 and 20 years was 98%, 88%, 77% and 45% respectively [20].There were no deaths in our LN group and all patients had preserved renal function at 12 months and 86% at 51 months of follow up.
Studies have shown a declining eGFR is a risk factor for poor outcome of systemic disease related nephropa-thy [2] [21].We have shown a clear trend of improvement in eGFR in vasculitis patients which correlates with a better renal and overall survival.Compared to results on LN from a single center cohort that showed a progressive decline in eGFR over a long term follow up of 25 years, our study showed a trend to improvement in eGFR in LN patients [22].This could however be attributed to a shorter period of observation in our study.The level of proteinuria is associated with the degree of renal scarring and poor long term renal outcome [23].Reduction in proteinuria is an important measure of favorable response to treatment, particularly in LN [24] [25].There was a decline in uPCR, a surrogate marker of renal improvement in patients with LN and vasculitis.
During our observation, there were no deaths in either vasculitis or LN groups.This could be attributable to a few factors.First, the period of follow up of 4 years is relatively short as compared to most published studies on patient survival in LN or renal vasculitis [2] [20]- [22] [26].Second, the numbers of patients were relatively small in both groups.Third in our study observations dated from the first visit to our clinic rather than from the time of diagnosis while most of our comparator groups have studied outcomes from the date of diagnosis.
In summary, our results show better patient renal survival, improvement in eGFR and reduction in uPCR compared to the published studies.Our findings from four years of observation support advocating combined renal rheumatology clinics in managing renal disease from systemic connective tissue disorders.

Figure 1 .
Figure 1.Profile plots for mean estimated glomerular filtration rate and mean urine protein creatinine ratio during follow up in vasculitis (a) and (b) and Lupus Nephritis (c) and (d).Time measured in 6 month intervals.

Table 5 .
GEE analysis for vasculitis and LN group of Patients eGFR over a time (0 a used for comparison variable).