To the Editor,
Lupus nephropathy (LN) has a direct impact on the survival of patients with systemic lupus erythematosus (SLE).1 Twenty percent of LN patients develop an end-stage chronic kidney disease 5-10 years after being diagnosed.2 Two phases have been established for treating LN: induction therapy and maintenance therapy.3 We present the case of a patient diagnosed with SLE, who had severe kidney damage. While the patient was undergoing LN induction treatment with steroids and cyclophosphamide, his kidney condition worsened, having developed a severe nephritic syndrome.
The patient was a 22-year-old man, smoker, with personal history of migraines and ventricular extrasystoles. He had been examined by the rheumatology and haematology departments due to thrombocytopenia. He was not allergic to any medication.
The patient attended the emergency department with a fever of >38ºC over a 10-day period, with asthenia and joint pain.
On physical examination, he was in a generally good condition, conscious and aware of his surroundings. Blood pressure was 120/80mm Hg; temperature: 38ºC; heart rate: 90 beats/min.
He had a bilateral malar erythema and mucocutaneous paleness. He had piercings and tattoos. He did not have jugular ingurgitation, and the rest of the examination was normal.
Biochemical analysis (blood): creatinine: 1.3mg/dl; uric acid: 7.2mg/dl; albumin: 1.7g/dl; total protein: 4.2g/dl; and the remaining parameters were normal. The haemogram showed: haematocrit: 31.3%, thrombocytopenia: 52 000, and an erythrocyte sedimentation rate (ESR) of 22mm/first hour.
In the systematic haematuria sample and white blood count, the 24-hour urine protein was 5.90g/24h and creatinine clearance was 71ml/min.
Immunological test had hypocomplementaemia C3: 45mg/dl (normal value [NV]: 79-152mg/dl); C4: 6mg/dl (NV:16-38mg/dl); normal immunoglobulins; and C-reactive protein: 1.23mg/dl (NV: 0-0.8mg/dl). The autoimmunity test: ANA +1/160 (speckled pattern); anti-DNA: >400; rest of antibodies, negative. Indirect Coombs test was negative. In the blood electrophoretic profile, hypoproteinaemia and hypoalbuminaemia were detected. The coagulation test was normal, with positive IgG and IgM anticardiolipins.
There were no relevant findings from the chest X-ray. We saw that the patient’s kidneys were 14.5cm and hyperechogenic in the kidney ultrasound.
The SLE diagnosis was made and a kidney biopsy was performed to assess the kidney damage: 40 glomeruli per cross-section were found and one of them was sclerotic. All of the glomeruli had a similar diffuse proliferative appearance, and more than 50% (36 glomeruli) were associated with an extracapillary proliferation (crescents). Fibrinoid necrosis was also observed in some glomeruli, as well as wire loops and haematoxylin bodies, with positive immunofluorescence: IgG (+++), IgM (+++), IgA (+++) and C3 (+++). No fibrosis or atrophy was found in the interstitial space, but mild non-specific lymphocytic infiltration could be seen. The activity index was 18/24 and chronicity 1/12. The definitive diagnosis was lupus nephropathy type IV.
Given the biopsy findings, treatment with 1g i.v. of 6-methylprednisolone (three pulses), followed by prednisone 1mg/kg/day p.o. and 1g cyclophosphamide. The patient was discharged and monitored on an outpatient basis.
After being discharged (14 days after having received the first cyclophosphamide pulse), the patient attended the emergency department once again with joint pain for 2-3 days, oedemas in the lower limbs and facial oedema; gross haematuria (coca-cola coloured), and fever. The patient said that he had completed the treatment prescribed upon discharge and had no other clinical findings that indicated infection.
On physical examination, he was in a generally good condition; he was conscious and aware of his surroundings. Blood pressure was 161/83mm Hg; temperature 36.5ºC; and heart rate 95 beats/min. He still had bilateral malar erythema and mucocutaneous paleness. During the rest of the examination we found oedemas on the lower limbs up to the top of the thigh.
Biochemical analysis (blood): creatinine: 1.2mg/dl; albumin: 2.7g/dl; total protein: 4.6g/dl. The haemogram showed that the haematocrit had decreased to 24.4%, and the thrombocytopenia remained at 51 000. The remaining parameters were normal.
The new sediment revealed macroscopic haematuria and haematic casts.
24-hour urine protein had increased to 31g.
In the immunological test we found a reduction in immunoglobulin G to 55mg/dl, with C3: 83mg/dl (NV:79-152mg/dl); C4: 1.6mg/dl (NV: 16-38mg/dl) and C-reactive protein: 0.12mg/dl. The autoimmunity test: ANA +1/80 (speckled), negative anti-DNA and cryoglobulins.
The chest X-ray and Doppler kidney ultrasound did not show any pathological findings. The blood culture, urine culture and joint fluid culture (knee) were all negative.
We repeated treatment with pulses of 1g of 6-methylprednisolone (three pulses) followed by prednisone at 1mg/kg/day and increased the cyclophosphamide pulse to 1.5g, which was brought forward. This decision was made because the patient presented with well-developed nephritic syndrome, he had lupus activity and had previously been diagnosed with type IV lupus nephropathy. We also considered that renal vein thrombosis, infectious processes and therapy non-adherence had already been ruled out. The patient then received pulses of cyclophosphamide 1.5g/month (6 months) and cyclophosphamide 1.5g (every three months, two cycles), with a tapering regimen of steroids. He was later administered a maintenance treatment with azathioprine.
With this treatment, progress was favourable. The patient was not hospitalised again, and kidney function normalised (plasma creatinine: 1.1mg/dl) and negative proteinuria. Although slight microhaematuria persisted, the haemogram and immunological test normalised.
Cyclophosphamide is considered to be the immunosuppressant with the best results in induction treatment of severe forms of proliferative LN.4 After having completed the induction treatment, it is possible that up to 20% of LN patients do not respond adequately to induction immunosuppression.5 In our case, less than 2 weeks had passed since the treatment had started to be able to consider the patient non-responsive. In non-responsive patients, compliance must be ensured, and renal vein thrombosis and infections ruled out.3 In our case, the patient had received cyclophosphamide 1g in the hospital (together with pulses of glucocorticoids i.v. and oral steroids, which the patient told us he had not stopped taking). We also ruled out infectious processes, as well as renal vein thrombosis. Nevertheless, systemic manifestations relapsed and his renal condition worsened, presenting with well-developed nephritic syndrome. For refractory cases, various therapeutic options are indicated, including: increasing glucocorticoid doses (repeating pulses of 6-methylprednisolone), repeating induction treatment with cyclophosphamide, using calcineurin inhibitors, mycophenolic acid, plasma turnover, depleting lymphocytes with rituximab.3 Recent meta-analyses deem induction therapy with mycophenolate more efficient than cyclophosphamide in severe LN.6 However, these results should be interpreted with caution given the limitations:3 Although most studies included severe LN, others also included other LN forms, i.e. type III and V. That is why, cyclophosphamide is the best immunosuppressant with the best result in severe LN relapse.4 In our case, given the severity of the process, together with the biopsy findings, we decided to repeat the regimen that was initially established at 1g i.v. pulses of 6-methylprednisolone and one pulse of cyclophosphamide i.v., increasing dosage to 1.5g in this case. After this moment, the patient’s progress was satisfactory. The American National Institute of Health (NIH)7 promotes a gold standard cyclophosphamide dose of 0.75-1g/m2, while the Euro-Lupus Nephropathy Trial2 states that 500mg every 15 days should be administered. In our case, we administered cyclophosphamide 1g at the start of treatment and in less than 15 days the patient presented systemic and renal activity again. Given the favourable response after increasing the cyclophosphamide pulse to 1.5g/month, it is possible that this lupus flare could have been due to the initial underdosing of cyclophosphamide.
To conclude, it is important to take into consideration the cyclophosphamide dosage administered before considering it to be inefficient and/or assess other alternative therapy regimens, as cyclophosphamide underdosing could be the cause of non-responsive patients.
