In a retrospective survey of 117 operations for primary or recurrent p
terygium, conjunctival autografting was compared with both excision wi
thout conjunctival closure ('bare sclera excision') and excision with
complete conjunctival closure. The probability of corneal recurrence a
t 36 months after surgery was determined by survival curve analysis. I
n previously unoperated cases conjunctival autografting (n = 15) resul
ted in a 14% probability of recurrence, compared with 70% for bare scl
era excision (n = 50) and 69% for excision with complete conjunctival
closure (n = 20). In previously operated cases conjunctival autografti
ng (n = 17) resulted in a 7% probability of recurrence, compared with
82% for bare sclera excision (n = 15). Hazard ratio analysis confirmed
the statistical significance of these results at the 95% confidence l
evel. Conjunctival autografting was more likely to produce an improvem
ent in visual acuity than other forms of surgery.