Bowel Preparation for Excretory Urography: a Randomized Trial - Research Paper Example

Critical Appraisal of the Article: Jansson, M., Geijer,H., & Andersson, T. (2007). Bowel preparation for excretory urography is not necessary: a randomized trial. British Journal of Radiology, 80: 617-624. Introduction Before conducting a detailed critical appraisal of the paper, some of the points which are worth mentioning as we look at the paper are: 1. There is a clear pathway of the patients who underwent the trial, those who were excluded and included and those in each arm of the study. 2. The description of the method, the criteria for outcome measures, the baseline characteristics of the patients, the trial profile, the results- all are depicted in tabular formats thus allowing easy understanding and comparison. 3. The outcome measures are very clear, i.e., quality of images according to European Commission image quality criteria for excretory urography. Title and authors The title of the study is "Bowel preparation for excretory urography is not necessary: a randomized trial." The title does reflect the content and also the method of the study as described below. All the 3 authors are radiologists with specialization in radiology, belonging to the Department of Radiology, Orebro University Hospital, Orebro, Sweden. Abstract The abstract is structured, short and to the point, summarizing the study and the main findings. "210 consecutive patients were randomized to three preparation groups with 70 in each group." The mode of randomization is very clear with this statement. However, the population studied is not identified well in the abstract. It is not clear whether those who were recruited hailed from which hospital and department. It is also not clear as to whether the patients who underwent the procedure are those with certain indications for urography or are those who have volunteered for the procedure. Introduction The introduction clearly presents the background of the study. The authors have clearly elaborated the importance of excretory urography and have questioned the need for bowel preparation prior to the procedure with examples from a comprehensive literature review. The purpose of the study is relevant and clearly stated. The latest articles discussing these issues, prior to the commencement of the study were studied and compared with. The research question asked in this study, i.e., is there a need to prepare the bowels before excretory urography follows on logically from this. Objectives and hypothesis The objectives are clearly stated. The intervention is described as "bowel preparation for excretory urography." The hypothesis is that bowel preparation is not necessary for excretory urography, which means that the images obtained for the procedure without bowel preparation are as good as those with bowel preparation in terms of purgation and diet restriction. This study actually has 3 arms. The Study The study design is that of a randomized control trial. According to Lachin (1998), "RCTs are considered the most reliable form of scientific evidence in healthcare because they eliminate spurious causality and bias." This method of study design ensures that known and unknown confounding factors are evenly distributed between treatment groups. According to Beller et al (2002), it is important to randomize because, randomization aims to obviate the third possibility. Allocation of participants to specific treatment groups in a random fashion ensures that each group is, on average, as alike as possible to the other groups (Beller et al, 2002). This study included 210 patients due to undergo excretory urography on out-patient basis. They were recruited consecutively. The study was held between September 2003 and April 2004 at the Orebro University Hospital, Orebro, Sweden. The inclusion criteria were ambulatory patients and older than 15 years of age. Exclusion criteria were contraindications to laxatives, such as small bowel stoma, colostomy and previous colon resection. Laxatives cause bowel purgation and can affect the results. The other 3 conditions can interfere with bowel dynamics and also can affect the results. The population studied, the inclusion criteria and exclusion criteria are clearly described under the study. To arrive at the number "210", the researchers initially included 232 patients of which 22 patients refused participation with only 210 patients participating in the end after consent. The need for follow up is not important in this study because the intervention study is only for the image quality of the pictures taken and not for the study of its effect on patient. There is no mention of the side effects of the medication used for contrast and this is also not important because the authors are not interested in studying about the contrast medium. Randomization was initiated after consent and ethics committee approval. The 210 patients were allocated to 3 groups equally (70 in each group). The 3 groups were: Group 1 (Bowel purgation together with dietary restrictions): This group received standard preparation consisting of 2 litres of polyethylene glycol electrolyte solution after 4 hours of starvation in the afternoon the day before excretory urography. Group 2: (Dietary restriction alone): This group was instructed to fast for 12 hours before excretory urography. Group 3: (No preparation): This group did not receive any preparation at all before excretory urography. This is a good-size study. The design of the study is so simple that other interested researchers can easily reproduce the study. The sample size was based on power analysis with the power set to 90% at a 5% significance level. This resulted in 44 participants per group, a total of approximately 150 patients. This study included 176 patients. The relevant subjects were clearly described. Randomization was performed according to the Consolidated Standards of Reporting Trials statement. High quality randomization will be characterized by a method of allocation that is not open to manipulation or statistical artifact. Randomization was concealed. According to Beller et al (2002), "it is very important that those responsible for recruiting people into a trial are unaware of the group to which a participant will be allocated, should that subject agree to be in the study. This avoids both conscious and unconscious selection of patients into the study.” Blocked randomization was used. Permuted block randomization ensures treatment group numbers are evenly balanced at the end of each block (Beller et al, 2002). Equal size of groups and distribution of patients was done. The examining radiographer was unaware of the patients preparation group. The 3 radiologists who examined the films were unaware of the patients preparation group assessed all images independently. Patients were analyzed in the groups to which they were randomized. The demography profile included age, sex and weight of the patients. The quality of the images was analyzed. The analysis was done with an intention to treat. All the groups were shown to be similar in all known determinants of outcome and there was no need for analyses to be adjusted. The 3 groups did not begin to differ in terms of the primary end point. Neither the patients, nor the examining radiographer or the radiologists were aware of group allocation initially. However, once the study started, the patients were aware of their group allocation but not the radiographer or the radiologists. It is extremely difficult to establish complete blinding to all parties. Equally important, however, is whether those doing the analysis were aware of the identity of either group. Somewhat typically these details are not given. The intervention used in all the 3 groups was similar, i.e., excretory urography. After the initial survey images of the abdomen were obtained, the contrast medium iohexol (300 mg/ ml) was administered intravenously. The dose was uniform for all the three groups, i.e., 40 ml in all patients weighing under 80 kg and 50 ml in those weighing above 80 kg. After this, standard nephrographic and pyelographic effect images were then obtained, first with and then without ureteral compression. The images were obtained with a flat panel detector mounted in a bucky table and integrated into a picture archiving and communication system. All images were evaluated on the same type of monitor in order to avoid variations. Zooming and alteration of the grayscale were allowed. These were compared between all the three groups. The outcome measures of this study were quality of images according to European Commission image quality criteria for excretory urography. This was compared between the groups using equivalency. Equivalency was defined as a difference not larger than 0.5 points in European Commission score where the maximum score was 9. Potential source of bias Bias may be defined as delivery of opinions, values and views as if they were universal without looking at other points of view. In this study, the researchers did not know how many patients did in fact follow our recommendation concerning preparation. They did not even attempt to ask the patients if they followed the instructions properly. This is important because, the patients were ambulatory and not inpatients. The authors did not question the patients because they did not want to create any further confusion. Construct validity Construct validity may be defined as the extent to which a scale measures a certain construct (Parahoo 2006) in which the link between conceptual definitions and operational definitions is examined. There is clear definition as to what is the intervention studied. External Validity and Generalizability Generalizability is a measure of the applicability of study findings and conclusions to other similar settings and populations (Holloway & Wheeler 2002). This aspect is important to know in the study because it gives an idea as to what extent the suggested causality is likely to hold true in other settings. The population studied represents general population of that country. Hence the findings can be generalized. Rigour and trustworthiness Rigour is the means to demonstrate integrity and competence and is determined by its truth value, applicability, consistency and neutrality (Holloway & Wheeler 2002). It involves rigid adherence to research designs as mentioned in the methodology of studies and precise statistical analyses and can be influenced by poor observations and failure to utilise the available data (Burns & Grove 1997). In this study, the authors maintained rigid adherence to the study design defined under methodology. The quality of data and the confidence of data display was gauged based on trustworthiness. Analysis and results Statistical analysis was done. Intention to treat analysis of assessments was performed, using the t-test for comparison between Groups 1 and 2 and Groups 1 and 3. Also, after this, all criteria were looked at separately and results were presented as proportions of patients in whom image criteria were judged as fulfilled when all three observers were in agreement. For proportions according to the binomial distribution, confidence intervals were calculated. Statistical analysis was done using computerized software. For discrete variables like residual gas and faeces, endpoint data were compared by means of chi square analysis. For normally distributed continuous variables, end point data were compared with Student’s t test. The results were briefly restated. The study proved that there is no need for bowel preparation for excretory urography. The results follow logically from the results and discussion. 210 patients participated in the study after consent. 70 patients were enrolled in each group. 34 patients dropped out after randomization. This resulted in 176 evaluated patients. The end points measured were the quality of images according to the European Commission criteria for excretory urography. No other outcomes were looked for. The tendency to be overwhelmed by the sheer number of outcomes being reported has been alluded to earlier. There was no statistically significant difference between groups after analysis of all criteria. A criterion was considered to be fulfilled only when all three observers were in agreement. The results of the study showed that fulfillment of the European Commission criteria was equivalent in the three preparation groups. The authors reported that bowel preparation before excretory urography is unnecessary and should be abandoned. It can be said that the analysis was done using appropriate statistical methods and hence the results are valid. References Beller, E.M., Gebski, V., Keech, A.C., (2002). Randomisation in clinical trials. MJA 2002, 177 (10): 565-567 Burns, N., Grove, S. K. (1997). The Practice of Nursing Research: Conduct, Critique and Utilisation. 3rd edition. Philadelphia: W. B. Saunders Co. Holloway, I., Wheeler, S. (2000). Qualitative Research in Nursing. New York: Wiley-Blackwell Lachin, J.M., Matts, J.P., Wei, L.J. (1988). Randomization in Clinical Trials: Conclusions and Recommendations. Controlled Clinical Trials, 9 (4): 365-74 Parahoo, L., 2006. Nursing research: Principles, Process, and Issues. London: Macmillan. Read More