The use of hydroxyurea and postoperative heparin prophylaxes were also included in the recorded observations made by the researchers. Standard statistical techniques including regression analysis were used to determine the levels of correlation that existed between transfusion therapy, pharmaceutical use, and complications. Materials

As this was a retrospective study that did not involve any direct experimentation or even direct observation, the materials used were rather minimal compared to many other clinical surveys. Patient records formed the entirety of the research materials used; these were accessed in accordance with privacy laws and ethical standards for the use of private medical information, with the records rendered anonymous prior to primary research being conducted and with full hospital permission. Records were accessed primarily electronically, though hard copies of the seventy-five case histories actually used as subjects in the study were also obtained after having been sanitized of identifying information.

Materials used in analysis were equally minimal, as the statistical analysis performed in this retrospective manner required only a computer terminal and the software necessary to perform statistical analysis after information from the seventy-five included case histories had been input as data. Standard software was used for this purpose, with the bulk of the analysis conducted using the latest available version of the SPSS software mist typically used for statistical analysis in such cases. Encryption of data was utilized to some degree, though to a fairly minimal extent as there was not considered to be a high risk of unauthorized access nor of potential damage caused by the unauthorized release of any of the data used in the study, as no personally identifying information was entered into the software program alongside the individual data points that formed the primary data for this research.

Discussion of Tables

The current literature shows a great deal of disagreement as to the benefits, if any, of various forms of perioperative and preoperative procedures as a means of reducing postoperative complications in patients with sickle cell disease undergoing surgery (Al-Samak et al. 2008; Buck et al. 2005; Hirst & Williamson 2001; Vichinsky et al. 1995; Haberken et al. 1997). It was hoped that this research would help provide more conclusive results as to the efficacy of various preoperative and perioperative treatments and their effects on postoperative outcomes, despite the relatively small sample size and the use of only a single medical institution in the retrospective analysis. As might have been expected based on previous results of similar research, however, this did not prove to be the case, and the results of this research as discussed below remain fairly inconclusive in these regards.

The preoperative and perioperative treatments that were specifically examined in this study were the administering of heparin prior to surgery, the administering of hydroxyurea as a standard course of medication and/or prior to surgery, and transfusions -- whether simple transfusions or exchange transfusions -- prior to and during surgery. Measures of postoperative complications that were examined in this research included levels and rates of postoperative fever, vaso-occlusive crises, acute chest syndromes, and length of hospital stay. Information regarding basic demographic, the type of surgery performed in each case, and the preoperative appearance of vaso-occlusive crises and acute chest syndromes was also collected, but was not analyzed for correlation.

The subjects in the study had a mean age of twenty-four-and-a-half years; surgery durations had a mean of sixty-seven minutes with a standard deviation of approximately forty-eight minutes, and mean hemoglobin concentration was measured...

Even from these preliminary statistics, it can be seen that little statistically significant variation in hemoglobin levels existed regardless of pre-, peri-, or postoperative treatment measures. Over thirty percent (30.7%) of subjects received no transfusion as a part of treatment, with approximately one-quarter each of subjects receiving a top-up transfer (26.7%) or a complete exchange transfer (25.3%) and the remaining 17.3% receiving a partial exchange transfer. Approximately ninety percent of subjects had experienced a vaso-occlusive crisis or acute chest syndrome prior to surgery (89.3% and 92.0%, respectively).
Table 1 provides a comparison of different subject groups divided by transfusion type received and postoperative outcomes. There do appear to be slight correlations between transfusion type and rates of postoperative vaso-occlusive crises and postoperative acute chest syndrome experiences, but these correlations are not strong enough to be entirely reliable given the sample size nor are they adequately supported by the methods of analysis employed. Length of stay showed the greatest degree of variation between groups, yet information regarding this variation is also inadequate for a definite correlation to be stated.

Table 2 shows a comparison of these same groups and their postoperative outcomes as correlated with hydroxyurea and heparin treatments, and the results are show even less statistically significant differences than appear in table one. There might be a slight reduction in postoperative occurrence of fever when hyrdoxyurea is used as a preoperative treatment, but again the correlation is not strong enough to derive a definitive conclusion from the available data. Levels of vaso-occlusive crisis actually show a slight rise in patients treated with heparin, as well, though again not to a statistically significant degree, meaning that even if one were to conclude (prematurely and erroneously) that hydroxyurea conferred an advantage of limiting fevers following operations, this benefit would be certainly outweighed by the dangers presented. Heparin had even less effect in all areas than hydroxyurea.

Table 3 shows the correlation between hemoglobin levels after transfusions and the presence of postoperative complications. There does appear to be a negative correlation here, meaning that the higher the levels of hemoglobin measured following a transfer, the less chance there was of there being a postoperative complication. This was less true for postoperative acute chest syndromes, and there was actually a very slight positive correlation with the duration of postoperative hospital stays -- those with higher hemoglobin levels stayed slightly longer -- but again this is not very statistically significant. Even the relationships that do appear to exist to a significant degree must be backed up by further research before they can be stated with confidence.