A patient experienced phrenic nerve paralysis after doxycycline sclerotherapy for treatment of chylous fistula at our institution. The purpose of this study is to use physiologic testing to determine whether doxycycline is capable of inducing defects in neural function. A nonrandomized, controlled trial was performed with nerve-conduction studies to determine possible deleterious effects of doxycycline sclerotherapy. Thirty-eight CD rats were used and separated into four groups. Doxycycline was applied to the sciatic nerves of rats by either topical application directly on the nerve or by intraneural injection. Nerve-conduction studies were done before surgery and at 1,7, and 21 days after surgery. The results showed a statistically significant decrement in nerve-conduction velocity and strength of transmitted impulse in those nerves injected with doxycycline solution. Complete nerve block was seen frequently. This effect was not seen with topical application of doxycycline or normal saline solution or with intraneural injection of normal saline solution. This study demonstrates that doxycycline can induce a marked decrement in neural function when applied to the subepineural layers of the sciatic nerve in the rat. Therefore doxycycline sclerotherapy should be used with great caution in situations in which it could become exposed to nerves that have sustained surgical trauma.