Background/Purpose

To compare the risk of serious infections of immunosuppressive medications used for the treatment of lupus nephritis.

Methods

We performed an up to date systematic review and network meta-analysis  (NMA) by performing an updated search for randomized trails of immunosuppressive medications for lupus nephritis up to September 2013 with the help of Cochrane and ACR librarians. We updated the data from the systematic review for the 2012 ACR lupus nephritis treatment recommendations and the published Cochrane Review on lupus nephritis. We abstracted data related to infections from these trials. Bayesian network meta-analyses (NMA) were conducted. A binomial likelihood model, which allows for the use of multi-arm trials was used. Informed priors were assigned for basic parameters and odds ratios, as well as risk ratios and risk differences, and 95% credible intervals were modeled using Markov chain Monte Carlo methods. Brooks-Gelman-Rubin plots were used to assess model convergence. Model fit was examined using the deviance information criterion (DIC) and the residual deviance. The degree of inconsistency was assessed by comparing statistics for the deviance and deviance information criterion in fitted consistency and inconsistency models. In further sensitivity analyses, fixed effects models and models using vague priors were also conducted.

Results

31 RCTs with 2, 442 patients provided data.  There were twenty-five 2-arm, five 3-arm and one 4-arm trial. We found that tacrolimus was associated with significantly lower risk of serious infections compared to prednisone, cyclophosphamide, mycophenolate mofetil and azathioprine with a risk approximately one-third  (Table 1).  We also found that MMF-AZA (MMF followed by AZA) was associated with significantly lower risk of serious infections as compared to low dose CYC, high dose CYC or high dose prednisone, although this was based on fewer data (Table 1).  Other differences between immunosuppressives did not reach statistical significance. 

Conclusion

Tacrolimus and MMF-AZA combination were associated with lower risk of serious infections  serious infections compared to other treatment options for lupus nephritis. These numbers can help patients make informed decisions about treatment options for lupus nephritis. 

 

Table 1. Comparison of various drugs for the risk of infections in patients with lupus nephritis showing statistically significant results
Treatment	Reference	Odds Ratio     (95% Credible Interval [CrI])	Relative Risk     (95% CrI)	Risk Difference %   (95% Crl)
TAC	PRED	0.30 (0.10,0.86)	0.34 (0.11, 0.88)	-8.97 (-15.82, -1.46)
TAC	CYC	0.34 (0.13,0.87)	0.38 (0.14, 0.88)	-7.47 (-13.89, -1.26)
TAC	MMF	0.35 (0.14,0.82)	0.38 (0.16, 0.84)	-7.23 (-15.34, -1.53)
TAC	AZA	0.28 (0.09,0.81)	0.32 (0.11, 0.83)	-9.61 (-20.74, -1.73)
MMF-AZA	CYC LD	0.09 (0.01,0.80)	0.12 (0.01, 0.83)	-17.42 (-42.84, -2.08)
MMF-AZA	PRED HD	0.03 (0.00,0.59)	0.06 (0.01, 0.64)	-37.87 (-82.60, -3.54)
MMF-AZA	CYC HD	0.07 (0.01,0.56)	0.09 (0.01, 0.62)	-22.74 (-46.47, -6.14)

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-systematic-review-and-network-meta-analysis-of-the-risk-of-serious-infections-with-immunosuppressives-for-lupus-nephritis/