Purpose: To assess the long-term effect of ulnohumeral arthroplasty and the relationship between radiological recurrence and clinical outcome. Materials and Methods: Eleven elbows with primary osteoarthritis were analyzed at an average of eighty months after ulnohumeral arthroplasty. All patients were male with a mean age of fifty years. The outcomes were assessed using the Mayo Elbow Performance Score(MEPS) and the clinical and radiological results were compared. Results: Four elbows were not painful and six were mildly painful but one was not changed. The mean gain in extension was $15^{circ}$, in flexion $10^{circ}$. There were satisfactory results in 8 elbows(73%) and the mean MEPS was 81 points. All of eleven elbows had some degree of recurrent osteoarthritis and there was no correlation between radiological recurrence and clinical endpoints in nine elbows. But in two elbows, it appeared that recurrence of osteophyte at coronoid process was severe and caused fair outcome. Conclusion: Ulnohumeral arthroplasy is one of the effective treatment options for primary osteoarthritis of the elbow. The radiological recurrence did not correlated with the clinical outcome in most cases.