Results
As part of the presented research, secondary analysis was conducted of patient data provided by NIS and HCUP with a range of variables for the purpose of establishing correlations between the perfusion modality applied in cardiopulmonary bypass (pulsatile and non-pulsatile) and patients’ background and recovery information. The data provided by NIS is considered reliable and sufficient for the presented research, as it has been established that the new sampling and data collection methods that have been adopted by NIS since 2012 allowed refining the analysis of hospitalization and reducing the margin of error by approximately 45 percent (see Research Design and Methodology). NIS estimations of access to health care, health care costs, quality, population-wide outcomes, and other rates have been used in health care studies as well as policy making.
One of the advantages of using NIS data for the presented research was that it covered all the patients whose health care utilization and outcomes information could have been submitted by hospitals, i.e. patients covered by social health care programs, national social insurance programs, and private insurance, as well as uninsured patients. This scope allowed estimating health care outcomes on a broader scale and with increased reliability because many other sources of healthcare information might have overlooked large patient groups due to insurance issues, which is undesirable in a health care study because the type of insurance itself does not influence health care outcomes for a person.
Since the goal of NIS was to collect as much patient information as possible, populations were included that are often missing from other samples (such as Medicare Advantage patients), which is relevant to the results of the presented research because cardiopulmonary surgeries may be given to a patient in any population, and exclusion of any patient groups (except for young people who initially were not planned to be in the sample) would have been deemed undesirable.
Two considerations were chosen for this research to estimate health care outcomes and recovery successfulness of patients after a cardiopulmonary bypass: survival rate and length of hospital stay. The two groups of patients compared by these criteria were those who received pulsatile perfusion during their bypass and those who received non-pulsatile perfusion. For sampling, four criteria were chosen for selecting patients for the presented research from the patient data presented by NIS: first, the sample included patients with cardiopulmonary diseases; second, the patients in the sample were older than 65 years old; third, it included patients who were diagnosed with a disease that required a cardiopulmonary bypass surgery; and fourth, it excluded patients with chronic diseases, such as diabetes, renal failure, or hypothyroidism.
The exclusion of patients with serious chronic diseases was justified by the fact that their medical conditions might have affected their treatment outcomes and recovery in a way that could not have been attributed to the use of either perfusion modality during their bypass. This consideration of previous medical conditions is a major component of research design (see Introduction).
Also, it was decided to take into consideration, within the framework of the presented research, certain patient background information, including gender, age, and race, which was supplied by NIS along with the length of hospital stay and survival rate indicators. Statistical analysis software (SPSS) was used to analyze the connections among patients’ background data and their health care indicators relevant to the presented research.
All the patients in the sample designed according to the criteria described above were further divided into two groups: having received pulsatile perfusion and having received non-pulsatile perfusion. Comparing the length of hospital stay and the survival rate for these two groups was the main purpose of the presented research. The perfusion modality, the length of hospital stay, and background information (age, gender, and race) were taken as independent variables in the statistical analysis, and survival rate was taken as the dependent variable.
The two criteria of health outcomes are relevant to the presented research because both of them reflect how successful a patient’s recovery was. Higher mortality rates in one of the patient groups in the sample (either pulsatile perfusion or non-pulsatile perfusion receivers) were considered to be an indication of a less successful cardiopulmonary surgery, which can be particularly attributed to the perfusion modality during the bypass. Similarly, longer hospital stay is an indication of slower recovery, which may be the result of vital organs recovery, which is directly linked to the perfusion modality during a bypass (see Literature Review).
The presented results, therefore, fully address the research question about the benefits of pulsatile perfusion in comparison with non-pulsatile perfusion in terms of physiological characteristics and recovery indicators.
Before comparing the two main criteria of the presented research (the survival rate and the length of hospital stay indicator) for the pulsatile and non-pulsatile groups of patients, their background information was analyzed with the consideration of frequency and in connection to the two main criteria. The tables below show frequency for three age groups (see table 1), seven race groups (see table 2), and two gender groups (see table 3) and present relevant data on the length of hospital stay (see table 4) and mortality in hospitals (see table 5) upon the analysis of the presented research.
Table 1. Frequency for Age Groups.
Table 2. Frequency for Race Groups.
Table 3. Frequency for Gender Groups.
Table 4. Hospital Stay.
Table 5. Died in Hospital.
The analysis of data presented above enabled further analysis with the consideration of the background information in connection to the perfusion modality used in a bypass (see table 6).
Table 6. Characteristic.
Further, correlations were analyzed on the basis of obtained data (see table 7).
Table 7. Correlations.
N=5905; for correlations equal to or greater than 0.034, p<0.01; for correlations equal to or greater than 0.047, p<0.001.
- 1=died; 0=alive
- 1=male; 2=female
- 55=Pulsatile; 50=Non-pulsatile
- 1=White; 2=Black, 3=Hispanic, 4=Asia/Pacific Islander; 5=Native Ameican; 6=Other.
Finally, the obtained data enabled the analysis that is the main purpose of the presented research: the comparison of the survival rate and the length of hospital stay indicator of patients who received pulsatile perfusion during their bypass to the same two indicators of those patients who received non-pulsatile perfusion (see table 8).
Table 8. Hospital Stay and Died in Hospital for Pulsatile and Non-Pulsatile Perfusion.
The obtained results allow definitively affirming that those patients who had received non-pulsatile perfusion during their cardiopulmonary bypass surgeries stayed in hospital for a longer period that those who had received pulsatile perfusion, which indicated better recovery for the latter group. Also, among non-pulsatile perfusion patients older than 65 years old, almost ten percent died in hospital after their bypass, while among pulsatile perfusion patients, only approximately two percent died in hospital, which allows definitively affirming that the mortality among pulsatile perfusion patients is lower.
Discussion
If sampling and data collection are performed adequately and accurately, statistical analysis of the obtained data is a reliable tool for proving a hypothesis. The general hypothesis of the study was that pulsatile perfusion in cardiopulmonary bypass surgeries is a more beneficial modality in terms of patients’ health outcomes than non-pulsatile perfusion. However, there can be various ways of discussing potential benefits. For the presented research, a perspective was selected that regarded the beneficial effects of a perfusion modality based on two indicators of successfulness of postoperative recovery: the survival rate and the length of hospital stay.
It can be argued that these indicators do not fully reflect the successfulness of recovery, and this consideration constitutes one of the limitations of the presented research. It can be said that earlier discharge from hospital can be due to a patient’s initial medical condition—some conditions in which a cardiopulmonary bypass is required are more serious than other, which is why recovery from the surgery can take more time.
It can also be said that earlier discharge can be due to unidentified health problems, i.e. complications did occur but were not properly detected, which is why further studies might include the readmission rate in the statistical analysis of outcomes of pulsatile and non-pulsatile perfusion patterns. Similarly, mortality rate is an arguable criterion because the death of a patient can be caused by aggravations and complications that cannot be directly linked to a perfusion modality during his or her bypass, i.e. the initial state and the differences among patients’ conditions are necessary considerations in the use of both of these criteria.
Another major limitation of the research is the attempt to establish causality. The matter of translating correlations into causalities is one of the main difficulties in any research, and it should not be overlooked by a researcher. The observation of similarities in the dynamics of the development of two different phenomena does not necessarily indicate that the phenomena are linked (Greenhalgh 80).
In a generalized situation, more attention should be paid to the factors and forces that shape the dynamics of each phenomenon’s developments, and causality can only be confirmed if the connections among these factors or forces (of their identity) are established and proved. In the given case, it is definitively observed that certain indicators of recovery successfulness for two groups of patients are significantly different, but it is not an indication of the fact that the indicators are different due to the modalities that had been used to divide the patients into the two groups.
In other words, it can be argued whether patients who had received pulsatile perfusion during their bypass recovered sooner than those who had received non-pulsatile perfusion due to the perfusion pattern they received or due to other factors that were not considered in the presented research. Similarly, the causality between mortality and perfusion modality is not established definitively because the obtained results (see Results) did not describe the cause of a patient’s death, which is why it cannot be identified whether the death was caused by something the could have been affected by the use of a certain perfusion pattern of the heart-lung machine or by something else unlinked to the perfusion-related conditions of the cardiopulmonary bypass.
To address this complication, it should be noted that two criteria were selected for evaluating the successfulness of patients’ recovery instead of one specifically for the purpose of providing a more reliable correlation. The mortality rate can be higher for the non-pulsatile patient group, which, as it was established above, does not necessarily mean that a patient who received this perfusion modality in their bypass are more likely to die in hospital after this; and the length of hospital stay may be longer for these patients, which does not necessarily mean that a patient given non-pulsatile perfusion is likely to spend a longer period in hospital.
However, when both of these indicators show better health outcomes for one group—the pulsatile perfusion receivers—it is a more reliable indication of beneficial effects of this modality. In addressing the correlation-related limitations of the research, it is noteworthy that the P value (see table 8) was less than one-thousandth, which indicates that the results are statistically significant (Krzywinski and Altman 1041), i.e. it is extremely unlikely that there is no correlation between the perfusion modality and any of the two main criteria—the survival rate and the length of hospital stay.
Another possible doubtful aspect of the presented research is the inclusion of patients’ background information—age, race, and gender—in the sampling design and further use of these characteristics in the statistical analysis part of the study. It can be argued that patients’ background characteristics are not important because the purpose of the presented research was not to find correlations between these characteristics and the patients’ health outcomes in terms of postoperative recovery successfulness in connection to the perfusion modality used in those patients’ bypasses. Indeed, the statistical analysis did not generate any results confirming or disconfirming such correlations, but the inclusion of age, race, and gender considerations was still important. First of all, age was needed as a criterion in sampling because only patients older than 65 years old were analyzed in the presented research.
Further division into three age groups was needed to compare the recovery indicators among patients of different age, which is important because expected postoperative recovery indicators are significantly different for patients between 65 and 74 and for patients between 85 and 94 (in the latter category, the period of stay in hospital is expected to be longer, and the death in hospital is more likely).
Concerning gender, this consideration was needed to ensure that male and female patients experience similar effects of the use of one of the two perfusion patterns on their postoperative recovery. A different result, i.e. different effects on recovery for men and women, could have indicated that the impact of perfusion modality might be gender-specific. Similarly, the inclusion of race as a consideration in the statistical analysis is justified by the fact that the similarity of effects needed to be observed across race groups.
Overall, despite the fact that finding connections between patients’ race and gender and their recovery after a bypass with the use of a certain perfusion modality was not the purpose of the presented research, it was decided to include these considerations in the analysis in order to make sure that the sample is diverse and factors that may turn out important in the relationship between perfusion modality and recovery are not overlooked.
It should not be neglected that the presented research regarded two aspects of the effects of different perfusion modalities on patients’ health outcomes. One of the aspects, the correlation between a modality and indicators of postoperative recovery successfulness, was described in the Results section, as the data collection stage of the presented research was mainly dedicated to this aspect. However, another effect that the presented research pursued to identify is the relationship between perfusion patterns and physiological characteristics of a patient during and after a cardiopulmonary bypass surgery.
This effect was addressed in the Literature Review section, and it was demonstrated that many studies had found various positive effects pulsatile perfusion had on the physiological state of a patient exposed to a bypass. Particularly, it had been shown that blood flow in different vital organs, microcirculation, thyroid hormone homeostasis, and some other physiological characteristics of a patient were observed as more positive under the conditions of pulsatile perfusion than under the conditions of non-pulsatile perfusion.
However, the aspects of patients’ health outcomes related to postoperative recovery were addressed in the relevant literature, too, and similar results had been obtained by other researchers. Particularly, it was suggested that the use of pulsatile perfusion during cardiopulmonary bypass surgeries was capable of reducing the systemic inflammatory response syndrome, inotropic support, intubation period, and the period of stay in a hospital for a patient (see Literature Review).
At the same time, some studies had found that there had been no correlation between pulsatile perfusion and death rate, which contradicts the results of the presented study. By combining the two indicators of successful postoperative recovery and confirming that, according to both of them, pulsatile perfusion is more beneficial in terms of patients’ health outcomes, the presented research makes a solid contribution supported by statistical analysis to the modern understanding of the effects of different perfusion patterns in bypasses.
The main clinical implication of the presented research is that pulsatile perfusion has been confirmed to have a number of positive effects on patients’ health outcomes, including physiological characteristics and postoperative recovery, which is why this perfusion modality should be used more in cardiopulmonary bypass surgeries and should be preferred to non-pulsatile perfusion. Also, it is important that the results of the presented research, including findings in the Literature Review, should be explained to patients, as they are authorized to make the decision about the use of one or the other perfusion pattern in their bypasses.
However, when explained that pulsatile perfusion has been confirmed to decrease the length of hospital stay and the mortality rates among patients who received it, patients may ask their physicians why this is so. A physician can further explain that, according to some studies, pulsatile perfusion improves microcirculation and blood flow in vital organs, and these two factors, along with some other physiological effects, can contribute to faster and most successful postoperative recovery and to lower risks of death in hospital. However, it should be admitted that an actual answer—a complete and accurate one— to this question has still not been found. And this is the main direction for further research in the area of perfusion modalities in cardiopulmonary bypass surgeries.
First of all, future studies should be dedicated to identifying the effects of pulsatile and non-pulsatile features of natural blood flow in the human organism. These effects have not been separated, and it remains unclear how blood circulation is affected by each of the two modalities (see Introduction). These studies need to be conducted by researchers from different areas, including physiologists and fluid mechanics specialists. Also, an important area of further research is the effect of pulsatile and non-pulsatile perfusion in a bypass on vital organs. If it is scientifically confirmed that pulsatile perfusion is better for vital organs during a bypass than non-pulsatile perfusion, it will become an indisputable argument for using it exclusively in cardiopulmonary bypass surgeries and an explanation for the results of the presented research.
Works Cited
Greenhalgh, Trisha. How to Read a Paper: The Basics of Evidence-Based Medicine. John Wiley & Sons, 2014.
Krzywinski, Martin, and Naomi Altman. “Points of Significance: Significance, P Values and T-tests.” Nature Methods, vol. 10, no. 11, 2013, pp. 1041-1042.