Capillaries measured in canine diaphragm by two methods.

We have measured capillary distribution in costal and crural canine diaphragm using two methods: histochemical processing and perfusion fixation. Each of 18 dogs was deeply anesthetized, the abdomen opened, and the left inferior phrenic artery cannulated. The animal was heparinized and overdosed with pentobarbital. The right hemidiaphragm was frozen, either postexcision (Protocol 1) or intact with no preload (Protocol 2), for histochemical processing. The left hemidiaphragm was fixed by perfusion in situ using 2% glutaraldehyde, either with preload (Protocol 1) or without (Protocol 2). Costal and crural regions of each hemidiaphragm were sampled for analysis. Frozen samples were sectioned and processed for acid-stable (pH 4.0) ATPase activity; perfusion-fixed samples were postfixed, stained, embedded in Epon, and sectioned. Measurements were made using a digital imaging system. We found that muscle fibers had smaller cross-sectional areas in costal than in crural diaphragm; capillary-to-fiber ratio (C:F) did not differ by region and regional differences in capillary density could be attributed to differences in fiber size. Results depended critically on methodology. In perfusion-fixed muscle, fiber area was less, C:F was greater, and capillary density was greater than in histochemically-processed tissue. We conclude that capillary distribution is similar in costal vs. crural diaphragm and that perfusion fixation identifies capillaries more effectively than histochemistry.