Universal ureteral stent placement at hysterectomy to identify ureteral injury: a decision analysis

Transcription

1 DOI: /j x Gynaecological surgery Universal ureteral stent placement at hysterectomy to identify : a decision analysis MO Schimpf, a EE Gottenger, b JR Wagner c a Division of Urogynecology, Department of Obstetrics and Gynecology, Hartford Hospital, Hartford, CT, USA b Advanced Urology of South Florida, West Palm Beach, FL, USA c Department of Urology, Connecticut Surgical Group, Hartford Hospital, Hartford, CT, USA Correspondence: Dr MO Schimpf, 85 Seymour St Suite 525, Hartford, CT 06106, USA. mschimp@harthosp.org Accepted 24 March Published OnlineEarly 2 June Objective Iatrogenic during gynaecological surgery is associated with increased morbidity when not diagnosed during the initial surgery. Preoperative insertion of ureteral catheters may enhance intraoperative recognition of injury and repair, but it is controversial. We sought to analyse the costs of this approach. Design/setting/population A decision-tree analysis of clinical scenarios of using universal ureteral catheterisation compared with no catheterisation was conducted for benign abdominal hysterectomy and radical hysterectomy. Methods Diagnostic-Related Groups and Current Procedural Terminology coding and reimbursement information were used as calculated for Medicare patients in the USA. Main outcome measures Differences in projections of total hospital-related costs related to clinical scenarios of perioperative care for women undergoing hysterectomy with or without ureteral catheterisation. Results Universal ureteral catheterisation is cost saving when the rate of during benign abdominal hysterectomy or radical hysterectomy is greater than 3.2%. Conclusions The cost savings of universal ureteral catheterisation at hysterectomy depend on the injury rate but are minimal at common levels of injury. Keywords Cost-effectiveness analysis, decision-tree analysis, hysterectomy,, ureteral stent. Please cite this paper as: Schimpf M, Gottenger E, Wagner J. Universal ureteral stent placement at hysterectomy to identify : a decision analysis. BJOG 2008;115: Introduction Gynaecological surgery may account for at least 52% of iatrogenic. 1 3 Conditions such as endometriosis, large leiomyomata or pelvic inflammatory disease that may distort the normal pelvic anatomy may be associated with a higher risk of injury. 4 Both the anatomic distortion and the extensive surgery required in women with a gynaecological malignancy may result in a higher rate of injury in these cases. 1 6 Rates of during all hysterectomies for benign indications range from 0.4 to 2.5%, although the rate is higher in pelvic reconstructive surgery and gynaecological oncology Furthermore, when universal cystoscopy is performed, the ureteral complication rate is found to be higher by as much as five times, possibly because injuries are detected that would otherwise have been silent or detected postoperatively. 12 A recent study using universal cystoscopy confirmed a risk of 2.2% for during abdominal hysterectomy. 7 The rate of during radical hysterectomy is more difficult to estimate because fewer procedures are performed and there are more confounding factors. Most prior reports cite an injury rate of 0.7 3%, 6,8,13,14 but other papers cite up to 30%. 2,4,11 A higher rate of injury is to be expected in women who have previously received radiation therapy. 4,6 The question of universal ureteral catheterisation in gynaecological surgery is controversial with regard to prevention of. However, catheterisation may facilitate prompt recognition of during the initial procedure, which is associated with early repair and decreased morbidity. 11 Past reports have found that about 70% of ureteral injuries present postoperatively, which has been shown to require a statistically higher number of procedures to repair. 2,3,6,11,12 Cost-effectiveness analysis is a method used to model the costs and outcomes of specific interventions to define the relative monetary value of alternate treatments for a condition. Looking solely at financial aspects, this can help establish ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology 1151

2 Schimpf et al. the most effective care path to maximise the healthcare benefits within a set budget. 15 To provide additional information to the question of preoperative ureteral catheterisation to improve intraoperative diagnosis of in gynaecological surgery, we sought to analyse the cost information using a decision-tree analysis from the perspective of the payer for the hospitalisation regarding four options of perioperative care. Methods We selected two scenarios to analyse and created flow diagrams representing the clinical courses. The first was a total abdominal hysterectomy for a preoperative diagnosis of benign uterine leiomyomata (Figure 1) and the second was a radical hysterectomy performed for a preoperative diagnosis of malignancy (Figure 2). As the rates for during these two procedures differ, the two scenarios were chosen for analysis to provide information for either clinical situations. Relevant Diagnostic-Related Group (DRG) and Current Procedural Terminology (CPT) coding information were used as shown in Figures 1 and 2. Cost information pertains to Connecticut-based hospitalisation in the USA. If preoperative ureteral catheterisation was performed, weassumedthat99%injurieswouldbeidentifiedintraoperatively (Figures 1 and 2). This rate was based on consensus of the experience of the senior gynaecological oncologists, urologists and urologic oncologists at our institution. For the purposes of the sensitivity analysis, we allowed for 1% error or a range between 98 and 100% for the rate of diagnostic accuracy of preoperative ureteral catheterisation. DRG 359 (converts to 358 if ) CPT (plus codes as noted) ICD-9 procedure 68.4 (plus codes as noted) Diagnosis Ureteral injury Early repair of DRG 358 ICD Delayed repair of Endoscopic repair CPT DRG 358 and 305 4,795 ($9,401) 7,935 ($15,558) Pre-op stent CPT DRG 359 ICD , 59.8 No DRG 359 DRG 358 Open repair DRG 358 and 305 8,302 ($16,278) 3,712 ($7,278) Total abdominal hysterectomy (diagnosis: leiomyomata) CPT Early repair of DRG 358 ICD ,724 ($9,263) Ureteral injury Endoscopic repair CPT DRG 358 and 305 7,864 ($15,420) Delayed repair of No pre-op stents (patient discharged with percutaneous nephrostomy, readmitted) Open repair DRG 358 and 305 8,231 ($16,140) No DRG 359 3,641 ($7,139) Figure 1. Hospital cost projection for total abdominal hysterectomy for leiomyomata ± bilateral salpingo-oophorectomy ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology

3 Cost analysis of universal ureteral stents at hysterectomy (no conversion if ) CPT (plus codes as noted) ICD-9 procedure 68.6 (plus codes as noted) ICD9 diagnosis 180, 182 Ureteral injury Early repair of ICD Delayed repair of Endoscopic repair CPT and 305 5,717 ($11,209) 8,856 ($17,365) Pre-op stent CPT ICD , 59.8 No Open repair and 305 9,224 ($18,086) 5,105 ($10,009) Radical hysterectomy CPT Early repair of ICD ,646 ($11,070) Ureteral injury Endoscopic repair CPT and 305 8,786 ($17,227) Delayed repair of No pre-op Stents (patient discharged with percutaneous nephrostomy, readmitted) Open repair and 305 9,153 ($17,948) No 5,080 ($9,961) Figure 2. Hospital cost projection for radical hysterectomy for malignancy. When no ureteral catheters were placed, based on prior reports, we assumed that delayed diagnosis and repair occurred in 70% of women, while the other 30% were repaired immediately. 2,4,6,11 For the purposes of the sensitivity analyses, a range of 66 85% was chosen for the rate of delayed diagnosis and a range of 15 34% was chosen for the rate of intraoperative diagnosis and repair, based on our literature search. 1 4,6,9,11,13 For the purposes of this analysis, we assumed that the initial placement of the ureteral catheter, which would be performed by an experienced surgeon, would carry negligible additional risk, as supported by our literature search. 16,17 This was also found to be true in a study of temporary intraoperative ureteral stents during trachelectomy as performed by gynaecological oncologists, which cited no intraoperative stent complications. 18 When repair of was performed at the time of the initial surgery ( early repair in Figures 1 and 2), we assumed that the repair was an ureteroneocystotomy performed in the distal third of the ureter. 6,11 Because this is successful in more than 95% of women, we assumed for this analysis that all ureteral repairs were successful on initial attempt and no further procedures were required. 11 For purposes of this analysis, all women with a postoperative not recognised at the time of the initial procedure underwent an abdominal/pelvic computerised tomography scan with and without intravenous contrast for diagnosis. Past studies have shown that the average time to diagnosis may be as long as a mean of days, but we assumed for the purposes of this analysis that it was diagnosed postoperatively during the initial hospitalisation. 9,11 Costs of imaging tests and bloodwork were incorporated into the DRG ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology 1153

4 Schimpf et al. coding cost information from the hospitalisation. In women who sustained a, a percutaneous nephrostomy was performed and they underwent surgical repair during a separate hospitalisation several weeks later. Based on our review of the literature, typically, 60% of women require open surgical repair, which consisted of an ureteroneocystotomy for this analysis, and 40% can be treated endoscopically when recognition of a is delayed. 9 Endoscopic repair consisted of balloon dilatation or endoureterotomy in this analysis. 9 For the purposes of the sensitivity analyses, a range of 60 63% was chosen for open surgical repair and 37 40% for endoscopic repair based on published reports. 2,9 Cost information was obtained for each of the relevant DRG codes, which incorporated other costs associated with operative care, including anaesthesia costs, intravenous therapy costs, laboratory costs, nursing services fees, operating and recovery room charges, pharmacy costs and radiology costs (Table 1). Medicare reimbursement for surgeon fees were obtained from the hospital professional services office and added to the DRG billing information. Urologist surgeon fees were only included in the event that ureteral catheters were placed preoperatively or that a repair was required. For the radical hysterectomy scenarios, we assumed that an urologist was involved, even if a gynaecological oncologist might be able to insert ureteral catheters or repair a independently. A range of costs was obtained for the purposes of sensitivity analysis by using the highest and lowest costs paid by various insurance companies, although the Medicare costs remained the base-case scenario. All women were assumed to have no other medical or surgical co-morbidities that would ultimately affect coding for the procedure or hospitalisation. For benign hysterectomy, the DRG coding converts from 359 to 358 for cases in which a complication is encountered, which is reflected in the difference in costs seen in Figure 1. Given the extensive nature of radical surgery for malignancy, there is no difference in coding based on the presence of a complication, such as. Thus, the differences in the reimbursement seen in the decision-making tree in this situation are based on the benefits of early diagnosis and repair compared with delayed repair (Figure 2). Readmission Table 1. Hospital cost information by specific code DRG information Total average cost per patient hospitalisation*, ** as paid by Medicare (range for all insurance companies) (radical hysterectomy) 4, (range 2, ,656.65) $8, (range $4,055 22,856) DRG 358 (uterine/adnexal procedures for nonmalignancy with complication) 3, (range 1, ,329.10) $7, (range $3,574 12,410) DRG 359 (uterine/adnexal procedures for nonmalignancy without complication) 3, (range 1, ,837.24) $6, (range $3,589 7,524) DRG 305 (kidney and ureter procedures for non-neoplasm without complication) 3, (range 2, ,039.31) $6, (range $5,230 9,881) Surgeon fees (by CPT code) Medicare reimbursement information** Total abdominal hysterectomy (corpus and cervix), with or without removal of tube(s), ($994.03) with or without removal of ovary(s) Radical abdominal hysterectomy, with bilateral total pelvic lymphadenectomy and ($1,834.88) para-aortic lymph node sampling (biopsy), with or without removal of tube(s), with or without removal of ovary(s) Cystourethroscopy, with ureteral catheterisation, with or without irrigation, ($138.49) instillation or ureteropyelography, exclusive of radiological service Ureteroneocystostomy; anastomosis of single ureter to bladder ($1,109.55) Cystourethroscopy with insertion of ureteral guide wire through kidney to ($266.42)*** establish a percutaneous nephrostomy, retrograde Ureteral endoscopy through established ureterostomy, with or without irrigation, ($388.61) instillation, or ureteropyelography, exclusive of radiological service; with fulguration and/or incision, with or without biopsy *Does not include surgeon fees. **Uses information from fiscal year ***Uses information from fiscal year ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology

5 Cost analysis of universal ureteral stents at hysterectomy for delayed repair was coded as DRG 305, and costs were added to those from the original admission (Table 1). As we assumed, women were diagnosed during the initial hospitalisation, this code was only applied to women who required delayed repair and a second admission. Women who had a repair at the time of the initial surgery were not readmitted and, therefore, did not accrue additional readmission costs. Then, based on stated rates of in gynaecological surgery from the literature, three rates of injury were used to stratify the analysis. 1 9,13,14 For women undergoing hysterectomy for benign indications, these rates represent low frequency of injury (0.4% or P = 0.004), moderate frequency (1.5% or P = 0.015) and high frequency of (2.5% or P = 0.025). For radical hysterectomy, these rates represent low frequency of injury (0.7% or P = 0.007), moderate frequency (3% or P = 0.03) and high frequency of ureteral injury (30% or P = 0.3). In both cases, the moderate rate of injury was used as the base-case, or likeliest, scenario for the sensitivity analyses. To calculate the final cost for each arm of the decisionmaking trees seen in Figures 1 and 2, the costs were summed. Then, the probabilities were incorporated into the costs for the final analyses, including the probabilities of the various repairs being performed. These results are presented in Tables 2 and 3 for the two procedures. The amounts presented as difference represent the cost savings to the institution for each woman who undergoes a specific procedure. The costs or savings per year to an institution would be obtained by multiplying the cost difference between stented patients and unstented patients by the number of cases performed yearly at that institution. Sensitivity analyses were conducted for both hysterectomy scenarios, including one-way sensitivity analysis, a threshold sensitivity analysis and a Monte Carlo simulation using 5000 iterations. One-way sensitivity analysis and the Monte Carlo simulation were performed by varying each point in the decision-tree analysis, but threshold sensitivity analysis was only conducted varying the rate, as this was the most clinically relevant question. TreeAge Pro Suite 2006 (TreeAge Software, Inc., Williamstown, MA, USA), and Microsoft Excel 9.0 and Visio 2003 (Microsoft Corporation, Redmond, WA, USA) were used for calculations and creation of the clinical models. This study was exempted from Institutional Review Board review at Hartford Hospital. Results Cost information from fiscal year 2006 based on DRG coding, averaged per patient for each hospitalisation, is shown in Table 1. We used the Medicare information as the base-case scenario with ranges given for the highest and lowest insurance company information available at our institution. Surgeon reimbursement fees are also shown by CPT code with 2007 Medicare reimbursement figures in Table 1 for the relevant procedures. The dollar-to-pound exchange rate as of 1 January 2007 was used ($1 = 0.51). The cost information was summed for each arm of the flow charts shown in Figures 1 and 2. These figures incorporate the DRG codes as well as the surgeon fees for the procedures relevant to each clinical scenario. The difference in costs to the health system based on whether or not women are universally catheterised is shown in the last row of Tables 2 and 3. This analysis shows that universal ureteral catheterisation is not cost saving, except in the case of a radical hysterectomy when the rate is 3.2% or higher. The one-way sensitivity analysis, which was performed by varying each point in the decision-tree analysis, including costs, confirmed the robustness of our assumptions. No change in any variable using one-way sensitivity analysis produced any situation in which ureteral catheterisation became more cost saving than no catheterisation. There was a trend towards cost savings for total abdominal hysterectomy for benign indications. Threshold sensitivity analysis confirmed that at a 3.2% rate of during benign abdominal hysterectomy, universal ureteral catheterisation became minimally cost saving. This rate of was defined as the break-even point in the threshold analysis. Table 2. Total abdominal hysterectomy for leiomyomata Treatment Minimal rate of injury (0.4%) (total cost if injury and repair ) 1 (total cost if no injury ) Moderate rate of injury (1.5%) (total cost if injury and repair ) 1 (total cost if no injury ) Maximal rate of injury (2.5%) (total cost if injury and repair ) 1 (total cost if no injury ) No stents 3, ($7,166) 3, ($7,240) 3, ($7308) Stents placed 3, ($7,287) 3, ($7,311) 3, ($7333) Difference (2$121) (2$71) (2$25) The moderate rate of injury was used as the base case for sensitivity analysis purposes. All incorporate a frequency of 0.3 for early repair of if there are no stents and a 0.7 frequency for delayed repair. Women who underwent delayed repair were assumed to have endoscopic repair 40% of the time and an open repair 60% of the time. ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology 1155

6 Schimpf et al. Table 3. Radical hysterectomy for malignancy Treatment Minimal rate of injury (0.7%) (total cost if injury/any repair ) 1 (total cost if no injury ) Moderate rate of injury (3%) (total cost if injury/any repair ) 1 (total cost if no injury ) Maximal rate of injury (30%) (total cost if injury/any repair 3 0.3) 1 (total cost if no injury 3 0.7) No stents 5, ($10,001) 5, ($10,133) 5, ($11,681) Stents placed 5, ($10,107) 5, ($10,134) 5, ($10,452) Difference (2$106) (2$1) ($1,229) The moderate rate of injury was used as the base case for sensitivity analysis purposes. All incorporate a frequency of 0.3 for early repair of if there are no stents and a 0.7 frequency for delayed repair. Women who underwent delayed repair were assumed to have endoscopic repair 40% of the time and an open repair 60% of the time. Monte Carlo simulations demonstrated that the incremental cost of ureteral catheter placement was (±SD , 95% CI 1, ,154.64) ($61 [±SD $1,131.40, 95% CI $2,148 2,264]) per patient after incorporation of the frequencies and clinical possibilities of a benign hysterectomy. The mean cost per catheterised woman to the healthcare system was 3, (95% CI 2, ,249.83) ($6917 [95% CI $5,257 8,333]) and women who were not catheterised incurred a mean cost of 3, (95% CI 2, ,197.81) ($6,857 [95% CI $5,237 8,231]). For radical hysterectomy, the cost per catheterised woman to the healthcare system was 7, (95% CI 4, ,980.30) ($13,736 [95% CI $8,003 21,530]), whereas women without ureteral catheters accrued costs totalling 7,242 (95% CI 4, ,077.20) ($14,200 [95% CI $8,466 21,720]). The incremental cost per catheterised woman was (±SD 2,608.80) ($ [±SD $5,115.27]). The Monte Carlo simulations performed for both clinical scenarios favoured not routinely placing ureteral catheters for the purpose of cost savings. At a rate of 0.4%, which was the lowest rate found during our literature search, ureteral catheterisation added ($120) to the hospitalisation for each woman. This figure had declined to ($22) when the ureteral injury rate increased to 2.5%, which was a more common figure cited in the literature. The cost of an undiagnosed requiring readmission and delayed open repair is 1.72 times the cost of a detected injury in a woman who undergoes immediate repair. It is 2.29 times more expensive than the case of patient who undergoes surgery without stents and without a ureteral injury. Discussion Cost containment and quality assurance are becoming more important in medicine. 19,20 This analysis was intended to provide supplemental information for the question of the need for universal ureteral catheterisation. Hysterectomy is both the most common surgical procedure performed worldwide and the most common cause of ureteric trauma. 2 6,12 Thus, decreasing the costs and complications associated with this procedure could have a significant impact on the healthcare system. This analysis bears out that universal ureteral catheterisation does not save costs, except in the case of rates higher than 3.2% during benign abdominal hysterectomy or radical hysterectomy. Sensitivity analyses confirmed that this conclusion was robust to changes in our study assumptions. Ureteral injury during gynaecological surgery is a common problem and prevention is clearly desirable. The injury may be a crush injury, a cautery injury, a suture ligation, transection or may relate to postoperative oedema kinking off the ureteral lumen. The extensive dissection required in radical hysterectomy may result in devascularisation injury. 5,6,11 Over time, women with ureteral injuries may also develop fistulae involving the vagina, bowel or other pelvic organs. 4,9,13 Should a occur, early recognition facilitates immediate repair and the best outcome for the patient in this situation. 1,4 7,9,13 Delayed diagnosis can predispose the woman to pain, fever, infection, peritonitis, urinoma formation and permanent renal damage. 3,10 Our analysis suggests that these women also incur greater costs to the healthcare system, with hospitalisations costing 1.72 times greater than those with immediate detection of injury. Ureteral catheterisation has also been examined in women who had undergone colectomy, where the incidence of ureteral injury is %. 16,17 In one study in this population, the time required to place the catheters was approximately 11 minutes, which did not significantly increase operating room time, and there were no significant complications intraoperatively or postoperatively from catheter placement. 16 The colorectal surgery literature reports a % intraoperative complication rate from catheter placement, including perforation or laceration of ureter, infection, haematuria 1156 ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology

7 Cost analysis of universal ureteral stents at hysterectomy or anuria. 16,17,21 This information is difficult to translate to gynaecological surgery, given that malignancy or diverticulitis may be the reason for surgery as well as the complicating factor in stent placement. Additionally, these studies include both male and female patients and data are not stratified by sex. In another decision-tree analysis, universal cystoscopy with intravenous indigo carmine at the time of abdominal hysterectomy became cost saving at a threshold rate of 1.5%. 10 That our study found a threshold rate of 3.2% augments this finding, as ureteral catheterisation would be expect to come at additional cost, reflected by a different CPT code, and thus require a higher injury rate at which it is cost saving. Cost analyses necessitate making certain assumptions and therefore cannot be applied to all clinical possibilities. 15 The surgical repairs chosen for this analysis were based on the experience of the senior author (J.R.W.) and reflect those typically performed at our institution; the degree of laparoscopic capability at other institutions may differ. Surgeon preference for technique of repair may also differ between institutions and surgeons, but this analysis is only relevant for the procedures chosen because different procedures would incur different surgeon fees. Furthermore, as healthcare systems differ between countries, some of the specific financial information may not be relevant internationally. The general principles of this analysis, however, likely are relevant to physicians practising in other countries. We assumed that the ureteral catheters would be placed by an experienced urologist, and this option may not be available at every institution. We agree that there are inherent risks of ureteral catheterisation, including perforation, transient oedema and obstruction, but our literature search supported a low incidence of these complications. 16,17 Similarly, as this analysis looks specifically at universal ureteral catheterisation for abdominal hysterectomy, it cannot be generalised to vaginal or laparoscopic hysterectomy, whether supracervical or total, given the variable rates of ureteral injury with those procedures. While silent renal injury may occur with hysterectomy, the true incidence is unknown and could not be completely accounted for in this analysis. We also did not incorporate possible outpatient costs, including office visits, medications used at home, imaging studies or procedures, laboratory studies, loss of time at work or other indirect or intangible costs. In the postoperative global period, it is likely that some of these factors would not incur additional billable cost but would nonetheless require time and effort from an office staff and physician. Management at this point in a patient s care is also subject to a high degree of variability based on specific physicians and is difficult to otherwise standardise. Inclusion of these factors would increase the costs associated with women who were not stented and who ultimately require delayed repair. As the costs associated with these women are already the highest in the analysis, this may increase the relative cost savings of ureteral catheterisation. Additionally, we did not include costs associated with medical liability litigation, which may occur after an injury. Incorporating any of these costs, for which we did not have reliable, consistent data from state to state or internationally, is likely to increase the relative costs of the women who sustained an injury and possibly enhance the cost savings of ureteral catheterisation as well. The decision to perform universal cystoscopy or ureteral catheterisation with any hysterectomy procedure should be based on a number of factors. When thinking about costeffectiveness, surgeons should consider their personal ureteral injury rate overall in surgery and the risk any specific woman might carry based on her history or pathology. It is difficult to define the specific risks per patient that would correspond to the injury levels described in this analysis. Factors that would increase a patient s individual risk, such as endometriosis, prior pelvic infection, extensive prior pelvic surgery, malignancy or prior pelvic radiation, should be taken into account when considering surgery. Further research on this topic could next include a randomised trial of ureteral catheterisation during hysterectomy to determine the overall efficacy of the procedure. Additionally, prospective comparisons of universal ureteral catheterisation to less expensive and less invasive measures such as universal cystoscopy would provide valuable information for preoperative planning as well. Surgeons should also consider whether ureteral catheters help identify an injury and facilitate early repair or even prevent. Gynaecologists who do not have privileges to perform cystoscopy might find preoperative universal ureteral catheterisation in women at high risk for injury more time-efficient if a urology consultant would otherwise be required at unpredictable or multiple times during a case for cystoscopy or repair. Conclusions Based on these results, ureteral catheterisation should be considered for cost savings in women undergoing benign abdominal or radical hysterectomy in whom the risk of ureteral injury exceeds 3%. We believe that each surgeon should assess his or her personal rate and plan for ureteral catheterisation accordingly. Ethics approval Because this study involved use of publicly available information rather than any specific, patient-identifying information, this study was exempted from Institutional Review ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology 1157

8 Schimpf et al. Board review at Hartford Hospital, Hartford, CT, USA. The authors have a letter from the IRB detailing this decision. Contribution to authorship E.E.G. designed the analysis and provided an initial framework of the decision trees as well as the references and paper. M.O.S. revised and updated the decision-tree analysis and did the statistical analyses with the assistance of Dr Coleman (see acknowledgements). M.O.S. also updated, revised and expanded the text, figures and references of the paper. J.R.W. helped to conceive the project initially and subsequently provided senior expert guidance for the analysis, reviewed the appropriate codes and flow of the decision trees and reviewed all drafts of the paper and figures. Acknowledgements The authors thank Craig Coleman, PharmD, for his contributions to this manuscript. The study qualifies as exempt from IRB review at Hartford Hospital. j References 1 Dowling RA, Corriere JN, Sandler CM. Iatrogenic. J Urol 1986;135: Selzman AA, Spirnak JP. Iatrogenic ureteral injuries: a 20-year experience in treating 165 injuries. J Urol 1996;155: Ghali AMA, El Malik EMA, Ibrahim AIA, Ismail G, Rashid M. Ureteric injuries: diagnosis, management and outcome. J Trauma 1999;46: Neuman M, Eidelman A, Langer R, Golan A, Bukovsky I, Caspi E. Iatrogenic injuries to the ureter during gynecologic and obstetric operations. Surg Gynecol Obstet 1991;173: Mariotti G, Natale F, Trucchi A, Cristini C, Furbetta A. Ureteral injuries during gynecologic procedures. Minerva Urol Nefrol 1997;49: St Lezin MA, Stoller ML. Surgical ureteral injuries. Urology 1991;38: Vakili B, Chesson RR, Kyle BL, Shobeiri SA, Echols KT, Gist R, et al. The incidence of urinary tract injury during hysterectomy: a prospective analysis based on universal cystoscopy. Am J Obstet Gynecol 2005; 192: Liapis A, Bakas P, Giannopoulos V, Creatsas G. Ureteral injuries during gynecological surgery. Int Urogynecol J Pelvic Floor Dysfunct 2001;12: Giberti C, Germinale F, Lillo M, Bottino P, Simonato A, Carmingnani G. Obstetric and gynaecological ureteric injuries: treatment and results. Br J Urol 1996;77: Visco AG, Taber KH, Weidner AC, Barber MD, Myers ER. Costeffectiveness of universal cystoscopy to identify at hysterectomy. Obstet Gynecol 2001;97: Watterson JD, Mahoney JE, Futter NG, Gaffield J. Iatrogenic ureteric injuries: approaches to etiology and management. Can J Surg 1998;41: Gilmour DT, Das S, Flowerdew G. Rates of urinary tract injury from gynecologic surgery and the role of intraoperative cystoscopy. Obstet Gynecol 2006;107: Sharfi ARA, Ibrahim F. Ureteric injuries during gynaecological surgery. Int Urol Nephrol 1994;26: Angioli R, Penalver MA. Urinary tract injuries. In: Hurt WG, editor. Urogynecologic Surgery, 2nd edn. Philadelphia, PA: Lippincott Williams and Wilkins; pp Ubel PA, DeKay ML, Baron J, Asch DA. Cost-effectiveness analysis in a setting of budget constraints is it equitable? N Engl J Med 1996; 334: Nam YS, Wexner SD. Clinical value of prophylactic ureteral stent indwelling during laparoscopic colorectal surgery. J Korean Med Sci 2002;17: Chahin F, Dwivedi AJ, Paramesh A, Chau W, Agrawal S, Chahin C, et al. The implications of lighted ureteral stenting in laparoscopic colectomy. JSLS 2002;6: Abu-Rustum NR, Sonoda Y, Black D, Chi DS, Barakat RR. Cystoscopic temporary ureteral catheterization during radical vaginal and abdominal trachelectomy. Gynecol Oncol 2006;103: Eisenberg JM Clinical economics: a guide to the economic analysis of clinical practices. JAMA 1989;262: Siegel JE, Weinstein MC, Russell LB, Gold MR. Recommendations for reporting cost-effectiveness analyses. Panel on Cost-Effectiveness in Health and Medicine. JAMA 1996;276: Sheikh FA, Khubchandani IT. Prophylactic ureteric catheters in colon surgery how safe are they? Report of three cases. Dis Colon Rectum 1990;33: ª 2008 The Authors Journal compilation ª RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology