It should not be remembered as a de

It should not be remembered as a deterrent to the use of noninvasive tests that many hospitals have relied on successfully for a number of years.Dr.Comerota.I echo Dr.Wheeler’s comments.I think clinical observations differ significantly from objective analyses of patients who have clot harbored in the deep venous system and that is further emphasized by the data we have reported.However, I think it is important to realize that our data are reported on the basis of a single prephlebographic noninvasive test, and that noninvasive test sensitivity can be substantially improved by serial testing if indeed the initial test result is normal and clinical suspicion persists.We believe that noninvasive hemodynamic testing is particularly useful in those patients who have suspected deep venous thrombosis.Dr.Wheeler mentioned and called our attention to the excellent report of the clinical utility of IPG in Huisman’s study reported in the New England Journal of Medicine.As we see, in evaluating that study, these patients underwent selected ascending phlebography and no patient with a normal IPG had ascending phlebography.Fifteen percent of those abnormal IPGs were detected on subsequent or serial IPGs.Serial studies were remarkably absent in our data and must be considered.In addition, when one analyzes the results of ascending phlebography, even in the selected data by Huis- man et al.(reference 12)and selected phlebograms, we see that 89% of those patients had proximal DVT and only 11% had isolated infrapopliteal clot, which I think are remarkably similar to our diagnostic group.Dr.John J.Cranley(Cincinnati, Ohio).This is a landmark study.Dr.Comerota and his colleagues have delineated an area that many of us have surmised, but, to my mind, no one has proved previously.I accept their data completely, although with some disappointment.In our own setting, the PRG has been used diagnostically.During a 17-year period and with 769 confirmatory phlebograms, our sensitivity is 91% and our specificity is 94%, whereas, when we last reported our results authored by Dr.Comerota, they were 92% and 95%, respectively.Clinically, a patient with a swollen limb and an abnormal PRG has venous obstruction.If an extremity is swollen and the PRG is normal, you can be sure that the obstruction is not venous.If a patient has symptoms suggesting DVT, and the PRG is abnormal, the patient can be treated.If the test is normal, the test can be repeated until clinical suspicions are satisfied.The technique and interpretation of the test have not changed; however, these tests are functional in nature and we know now that DVT can exist without detectable functional impairment.We knew from the beginning that a clot outside the mainstream would not cause abnormal test results.We also knew that partially occluding clots would not cause abnormal results, but we did not know how frequently this occurs, particularly when detected by improved phlebographic and duplex scanning techniques.48 Comerota et ctl.We have learned that extensive chronic venous thrombosis may not cause functional impairment.Because of these findings, I have gradually modified my public statements concerning acute occlusive venous thrombosis.The functional tests have been and continue to be worthwhile and are very cost effective.This study teaches us that many partially occluding clots in the deep venous system exist and are not detectable by functional tests.The clinical importance of this is not known.I have long believed that any time blood can flow over a thrombus, that thrombus will lyse.Dr.Russell Hull(Hamilton, Ontario, Canada).During the past several years, the use of diagnostic tests(particularly the hemodynamic ones)and venous thrombosis have appeared to be somewhat of a riddle “wrapped around an enigma.”In other words, there has been much confusion about the clinical place of these tests in the screening of asymptomatic patients because of the lack of data about their overall accuracy when used for surveillance.This study emphasizes a key point that has been suspected for some time, that the overall accuracy of these tests for surveillance is inferior.Indeed, in patients having hip surgery, in the past we have believed that phlebography was required to determine whether prophylaxis was having an impact.In patients who have suspected venous thrombosis, the hemodynamic tests such as IPG or PRG are effective and safe to use for diagnosis.Paradoxically this may also be true for surveillance because there is a subtle difference in the trials evaluating diagnosis and surveillance.The subtlety goes back approximately 10 years, at which time the IPG and PRG were inaccurate for all thrombi because these tests missed calf thrombi.Several studies had to be done, the most recent one by Huisman and colleagues, to show that despite that inadequacy, when the test is repeated and results remain normal, the test