As early as 1968, blood transfusions used in cardiac surgery had been shown to be associated with risk of CMV infection, as evidenced by CMV serologic response; however, the incidence of reported CMV infection fluctuates between 40 percent in earlier studies and 1.5 to 12 percent in recent reports and, when considering only viremia, between 0 and 1 percent. In our experience, such an incidence is low; during the period of October 1985 through June 1986, we studied 30 outpatients with an uneventful course six weeks after cardiac surgery, and only one of them had viremia. In organ-transplant recipients, the incidence of CMV infection varies from 50 to 80 percent when combining serologic, culture, and histopathologic findings and is about 20 percent when considering viremia following either marrow (18 percent), renal (29 percent), or cardiac (20 percent) transplantation. Most research on CMV infection after cardiac surgery relies on serologic analysis. In our study, the laboratory diagnosis of CMV infection depended on isolation of the virus.
The sensitivity and precision of this approach is outstanding for viremia, but there have been conflicting reports on the reliability of CMV detection in urine. Our choicer of including urine cultures was guided by two considerations: (1) previous studies demonstrated a very low (less than 1 percent) rate of viruria due to CMV in normal adults; and (2) all six viruric subjects reported ill effects (atypical lymphocytosis, four; enzymatic abnormalities, six; jaundice, four; renal failure, four; and CMV-IgM presence, two) which were similar to viremic patients. Therefore, using the criteria of viral shedding, we have shown CMV infection in 29 of 115 patients suffering mediastinitis after open-heart surgery, for an incidence of 25 percent. Viremia was documented in 79 percent (23/29) of these patients. levitra professional
Although it may be thought that the frequency of CMV infection fluctuates, depending on immunologic disturbances observed in patients in the ICU, we were unable to find any published data that could substantiate such a statement. In our experience the rate of CMV shedding does not seem to be critically dependent on the severity of the patients status; for instance, it is important to underline that the indicator of severity of disease that we used, the APACHE II score, was the same (14.6 ±7.3) on the day of admission, whether patients subsequently proved to have CMV infection or not. Moreover, during the period of October 1985 through June 1986, only four of 36 patients who required prolonged care in our ICU (first-day APACHE II score: 16 ±8) after cardiac surgery for reasons other than mediastinitis were viral excretors. During the same period, 11 of 35 patients with mediastinitis evidenced viral shedding. There is a growing body of evidence for exogenous transmission of CMV even in seropositive patients, with the risk for acquiring viral infection linked to the number of transfused units. It is likely that our patients had received more blood products than the average currently required for standard ECC, because the duration of surgery is longer and reexploration for hemorrhage more frequent in patients who ultimately become infected. Although this warrants farther study, it is unlikely that the observed rate of viremia (20 percent) can be directly related to the number of units transfused. Unfortunately, we could not assess this point because of the number of referring surgical departments.