The current recommendations of kidney biopsy in childhood idiopathic nephrotic syndrome (CINS) were put forward to minimize unnecessary kidney biopsies in underlying minimal change disease (MCD). However, there remains a diversity of opinion about the criteria for biopsying children with idiopathic nephrotic syndrome. This study was conducted to prospectively study their usefulness in avoiding biopsies in MCD and to evaluate further modifications for minimizing biopsies in CINS. Of 400 consecutive CINS patients, 222 patients were subjected to kidney biopsy according to the current recommendations. The histopathology spectrum of these selectively biopsied children revealed focal segmental glomerulosclerosis (FSGS) in 39%, MCD in 34.2%, membranoproliferative glomerulonephritis (MPGN) in 16.2%, mesangioproliferative glomerulonephritis (MesPGN) in 7.6%, membranous nephropathy (MN) in 1.8%, and diffuse mesangial sclerosis (DMS) in 0.9%. We observed that despite the current recommendations and efforts to minimize biopsy, 34% of children had MCD on histopathology. Two or more clinical (hematuria and hypertension) or biochemical (renal insufficiency) parameters were present in all children with MPGN. Low C3 was present only in children with MPGN. All the steroid responders were found to have MCD, FSGS, or MesPGN on biopsy. Cyclophosphamide response correlated better with steroid responsiveness (P=0.02) than with histo- pathology (P=0.80) in MCD, FSGS, and MesPGN. Based on these observations, we suggest some modifications in current recommendations for kidney biopsy to minimize biopsying children with MCD. These are (1) biopsies in children (age 1–16 years) should be restricted (a) to a subgroup with two or more clinical and biochemical parameters and (b) in steroid non-responders, (2) the decision to administer cyclophosphamide should be based on steroid response pattern without requiring a prior routine biopsy.