Hepatic interstitial laser photocoagulation. An investigation of the relationship between acute thermal lesions and their sonographic images.
OBJECTIVES: The relationship between hepatic interstitial laser photocoagulation (ILP) lesions and their acute ultrasound images was evaluated. In addition, the natural history of ILP lesions in normal pig liver was documented. METHODS: Eighteen pigs underwent laparotomy and ultrasound-monitored ILP. In part 1 of the study, 12 pigs each had four separate exposures (1.50 W for 60, 100, 300, and 500 seconds) and were divided into four groups according to when they were killed (0, 3, 7, and 21 days). In part 2 of the study, six pigs each had two sequential exposures (1.60 W for 1,000 and then 500 seconds) at separate hepatic sites. Survival time was 3 days. Necropsy and histologic examination were performed in all animals. In 0- and 3-day survivors, actual thermal lesions were compared with "early" (immediately after ILP) and "late" (1 hour after ILP) ultrasound images. RESULTS: In the 300-, 500-, and 1,000-second exposures of parts 1 and 2, thermal lesions were overestimated or approximated by early ultrasound and were underestimated or approximated by late ultrasound. Analysis of variance showed statistically significant differences between thermal lesions and their early and late ultrasound images (F = 18.6, P < .001, no interactions). Time-growth characteristics of ILP lesions were reasonably consistent on ultrasound; exceptions were identifiable 200 seconds into the exposure. In part 2, ultrasound changes were minimal in five of six 500-second (second sequential) technically satisfactory exposures. Thermal lesions were seen at necropsy. All lesions healed by formation of granulation tissue and collagen. CONCLUSIONS: During ILP, early ultrasound images frequently overestimate actual thermal lesions. Ultrasound-monitored ILP of tumors may be most effective if, on early ultrasound, echogenic changes extend beyond the tumor margins. Late ultrasound images underestimate or approximate thermal lesions. Their value in clinical ILP should be investigated. It is unclear why ultrasound images of proven thermal lesions were not seen during 5 of 6 otherwise satisfactory 500-second ILP exposures performed immediately after 1,000-second exposures.