Does smoking cause nodular glomerulosclerosis?

Renal biopsy conference this past Thursday at the Brigham brought forth an interesting discussion about smoking causing nodular glomerulosclerosis. It was suggested that there was a strong association between smoking and glomerulosclerosis. So, what’s the bottom line? Is there a causal relationship?

One of the first substantive morphologic association study – and the one frequently cited - was by the group at Columbia, New York by Glen Markowitz published in Human Pathology.

Markowitz reviewed 5,073 native renal biopsy samples  at Columbia University between 1996 and 2001. They reported that after excluding diabetes mellitus, 0.45% biopsies or 23 patients had prominent diffuse and nodular mesangial sclerosis, glomerular basement membrane thickening, arteriosclerosis, and arteriolosclerosis. Patients with nodular sclerosis were mostly older white men and the majority had hypertension, renal insufficiency, hypertension,  and  proteinuria (nearly 70% had >3 g/d of proteinuria). Hypertension (95.7), smoking (91.3%), and hypercholesterolemia (90%) were highly prevalent. Follow-up data were available for 17 patients, 6 of whom reached end-stage renal disease (ESRD) (35.3%). By Kaplan-Meier estimates, the median time after biopsy to ESRD was 26 months. Predictors of progression to ESRD included continuation of smoking (P = .0165), lack of angiotensin II blockade (P = .0007), degree of tubular atrophy and interstitial fibrosis (P = .0517), and degree of arteriosclerosis (P = .0096). They concluded that idiopathic nodular was linked to hypertension and cigarette smoking.

There are several small studies that document an association. Liang et al from the Mayo presented a case in AJKD in 2007:

“In this report, we describe clinical and kidney biopsy findings for a 66-year-old woman with a history of long-term heavy cigarette smoking who developed proteinuria and decreasing renal function. This study also describes clinical and kidney biopsy findings for 9 patients with a history of smoking. None of these patients had hypertension, diabetes mellitus, or other risk factors that might result in vascular injury. Renal biopsy specimens showed a range of long-term changes with varying degrees of focal segmental or focal global glomerulosclerosis, nodular glomerulosclerosis, ischemic glomeruli, interstitial fibrosis and tubular atrophy, and mild to moderate arterial sclerosis and arteriolar hyalinosis. Electron microscopy often showed glomerular capillary wall thickening caused by subendothelial expansion by cellular elements and new basement formation resulting in segments of double contours. These changes indicate endothelial injury and glomerular capillary wall remodeling; the lesions mimic those seen in patients with chronic hypertension and chronic or healed thrombotic microangiopathies.”

Likewise Li et al in 2008 write:

“Idiopathic nodular glomerulosclerosis is an enigmatic condition closely resembling diabetic nodular glomerulosclerosis without evidence of diabetic mellitus or other specific disease. Idiopathic nodular glomerulosclerosis remains a rare disease entity with an unclear pathogenesis. Clinicopathologic features of 15 patients with idiopathic nodular glomerulosclerosis were evaluated in a retrospective review of renal biopsies between 1998 and 2007. Our study cohort consisted predominantly of older (mean age, 64.2 years) white (73%) women (67%). Fourteen patients (93%) had a history of hypertension, and 10 (67%) were active smokers at the time of biopsy. Nine patients (60%) were obese (body mass index, >30 kg/m(2)) and 4 (27%) were overweight (body mass index, 25-29.9 kg/m(2)). Fourteen patients (93%) presented with renal insufficiency with mean serum creatinine level of 2.8 mg/dL. All 15 patients presented with proteinuria (mean urinary protein excretion, 5.6 g/24 h). Eleven patients (73%) presented with nephrotic-range proteinuria and 8 (53%) with nephrotic syndrome. Histopathologic findings showed nodular glomerulosclerosis (100%), moderate to severe arterio-arteriolosclerosis (100%), and glomerular basement membrane thickening (100%). Immunofluorescence and electron microscopy studies had no other specific findings. Our results confirm previous studies of a close association of hypertension and smoking with idiopathic nodular glomerulosclerosis.”

Do these papers prove a causal relationship between smoking and glomerulosclerosis – no is the answer.