Incidence, Healthcare Resource Utilization, and Costs Associated with Intraoperative Ureteral Injuries: Evidence from a German Claims Database.

Why carry out this study? Intraoperative ureteral injury (IUI) during abdominopelvic surgery occurs infrequently but can cause significant clinical and economic burdens. This retrospective study examined the incidence of IUIs, all-cause healthcare resource utilization, and healthcare costs using claims data from a large, national German database between 2014 and 2021. What was learned from this study? Out of 159,581 patients who underwent an abdominopelvic surgery, IUI occurred in 217 patients (0.14%), with specific procedures like proctectomy and hysterectomy associated with higher incidence of IUIs. In comparison to patients without IUI, patients with IUI had significantly longer inpatient hospital stays and higher all-cause healthcare costs during the 1-year follow-up period; substantial variations were observed for gastrointestinal and gynecological (GYN) surgeries. Although rare, IUI poses notable risks with higher costs and extended inpatient hospitalizations, particularly for GYN surgical procedures, highlighting the need for risk awareness and potential preventive strategies.

Overall, 171,308 patients with abdominopelvic surgery at risk of IUI were identified. Of these patients, 159,581 (93.2%) were included in this study upon meeting the eligibility criteria. The moderate definition of IUI recommended by key external experts for its balanced sensitivity and specificity was therefore selected as the primary definition for the analysis. Baseline characteristics of patients with and without IUI before and after matching are shown in Table  1 . In the overall population (pre-matching), 217 patients were identified as having an IUI, and 159,364 did not have an IUI. Patients with IUI had a mean (SD) age of 59.8 (14.8) years and 71.9% were female, while patients without IUI had a mean (SD) age of 58.3 (16.7) years and 62.1% were female. Among patients with IUI, 95 (43.8%) underwent GI procedures, 92 (42.4%) GYN procedures, and 30 (13.8%) other abdominopelvic procedures; among patients without IUIs, 58.3% underwent GI procedures, 35.0% underwent GYN procedures, and 6.7% underwent other abdominopelvic procedures. Surgical procedures were performed in an inpatient setting for 97.7% of patients with IUI and 87.1% of patients without IUI. The proportion of patients with a cancer diagnosis was 29.5% among patients with IUI and 19.2% among patients without IUI. Table 1 Baseline characteristics of matched vs. unmatched patients with and without IUI Parameter Pre-matching Post-matching With IUI n  = 217 Without IUI n  = 159,364 With IUI n  = 216 Without IUI n  = 216 Age, years  Median (range) 60.0 (19.0–90.0) 59.0 (18.0–103.0) 60.0 (19.0–90.0) 60.0 (27.0–88.0)  Mean (SD) 59.8 (14.8) 58.3 (16.7) 59.8 (14.8) 59.9 (14.2) Sex, n (%)  Female 156 (71.9) 98,948 (62.1) 155 (71.8) 152 (70.4)  Male 61 (28.1) 60,416 (37.9) 61 (28.2) 64 (29.6) CCI score  Median (range) 1.0 (0.0–13.0) 1.0 (0.0–17.0) 1.5 (0.0–13.0) 1.0 (0.0–11.0)  Mean (SD) 2.3 (2.7) 1.8 (2.3) 2.3 (2.7) 2.3 (2.6) Patients with a cancer diagnosis, n (%) 64 (29.5) 30,651 (19.2) 64 (29.6) 61 (28.2) Surgical type, n (%)  GI 95 (43.8) 92,972 (58.3) 94 (43.5) 94 (43.5)  GYN 92 (42.4) 55,770 (35.0) 92 (42.6) 92 (42.6)  Other 30 (13.8) 10,622 (6.7) 30 (13.9) 30 (13.9) Surgical setting, n (%)  Outpatient 5 (2.3) 20,637 (12.9) 5 (2.3) 5 (2.3)  Inpatient 212 (97.7) 138,727 (87.1) 211 (97.7) 211 (97.7) CCI Charlson Comorbidity Index, GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury, SD standard deviation Baseline characteristics of matched vs. unmatched patients with and without IUI CCI Charlson Comorbidity Index, GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury, SD standard deviation Nearly all ( n  = 216) patients with IUI were successfully matched with patients without IUI for comparative HRU and cost analyses. After matching, baseline and clinical characteristics were similar among patients with and without IUI. Patients with IUI had a mean (SD) age of 59.8 (14.8) years, and 71.9% were female; patients without IUI had a mean (SD) age of 59.9 (14.3) years, and 70.4% were female. Incidence proportions of IUIs by type of abdominopelvic surgery are reported in Fig.  2 . Across all surgeries, the IUI incidence proportion was 0.14% (136.0 IUIs per 100,000 surgeries [95% CI 119.0–155.3]), with notable variation depending on the surgical type. The highest IUI incidence proportion was observed for “other” abdominopelvic surgeries (incidence proportion of 0.28%; 281.6 per 100,000 surgeries [95% CI 195.8–403.3]), followed by GYN surgeries (0.16%; 164.7 per 100,000 surgeries [95% CI 134.1–202.1]), then GI surgeries (0.10%; 102.1 per 100,000 surgeries [95% CI 83.4–124.9]). Among specific procedures, proctectomy-related surgeries (including partial proctectomy, total proctocolectomy, and proctectomy with anastomosis) had the highest incidence proportion of IUIs (1.03%; 1028.4 per 100,000 surgeries [95% CI 664.3–1576.4]) followed by hysterectomies involving removal of additional structures (0.74%; 739.0 per 100,000 surgeries [95% CI 507.3–1070.4]). Fig. 2 Incidence of IUIs per 100,000 surgeries in patients with abdominopelvic surgeries; descriptive analysis . Values are means with 95% CIs. a Including proctectomy, partial proctectomy, and total proctocolectomy (all with anastomosis). b Including subtotal or partial (segmental) colectomy with re-anastomosis. c Including excision (removal of a portion) of the uterus (including myomectomy). CI confidence interval, GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury Incidence of IUIs per 100,000 surgeries in patients with abdominopelvic surgeries; descriptive analysis . Values are means with 95% CIs. a Including proctectomy, partial proctectomy, and total proctocolectomy (all with anastomosis). b Including subtotal or partial (segmental) colectomy with re-anastomosis. c Including excision (removal of a portion) of the uterus (including myomectomy). CI confidence interval, GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury Among the GI surgical procedures analyzed, incidence proportions varied considerably. For example, the IUI incidence proportion of subtotal or partial colectomy with re-anastomosis was 0.22% (216.7 per 100,000 [95% CI 122.6–374.6]). GI tract surgeries had an IUI incidence proportion of 0.12% (118.9 per 100,000 surgeries [95% CI 82.1–171.4]), while other large intestine procedures had an incidence proportion of 0.02% (19.7 per 100,000 surgeries [95% CI 9.7–38.0]). Among GYN surgeries, IUI incidence proportions also varied substantially by procedure. Hysterectomy without additional removal of supporting structures had an IUI incidence proportion of 0.26% (259.8 per 100,000 surgeries [95% CI 172.6–388.3]). The IUI incidence proportion for patients with oophorectomy and/or salpingectomy was 0.19% (192.6 per 100,000 surgeries [95% CI 103.0–349.3]). Excision or removal of a portion of the uterus, including myomectomy, had an IUI incidence proportion of 0.09% (91.3 per 100,000 surgeries [95% CI 45.1–176.3]). Finally, retroperitoneal surgeries under the “other” abdominopelvic surgery category had an IUI incidence proportion of 0.28% (284.8 per 100,000 surgeries [95% CI 174.3–459.3]). All-cause HRU and healthcare outcomes at 1 year were compared among matched pairs of patients with and without IUI for all abdominopelvic surgeries ( n  = 216 matched pairs), and by surgical types with at least 46 matched pairs: GI surgeries ( n  = 94 matched pairs) and GYN surgeries ( n  = 92 matched pairs). Comparative analyses for “other” abdominopelvic surgeries were not conducted because of insufficient sample size. The matched pairs comparisons of HRU outcomes in patients with and without IUI for all abdominopelvic surgeries, GI surgeries, and GYN surgeries are shown in Table  2 . In general, higher HRU was observed among patients with IUI compared with patients without IUI. For all abdominopelvic surgeries, the mean hospital stay duration was significantly longer among patients with IUI than among those without IUI (19.1 days vs. 9.4 days; P  < 0.001). In addition, postsurgical inpatient admissions were significantly more common for patients with IUI than for patients without IUI (81.5% vs. 41.2%; P  < 0.001). Similarly, patients with IUI experienced a higher number of hospital admissions compared with patients without IUI, averaging 2.5 vs. 1.0 ( P  < 0.001). Table 2 Propensity score-matched analysis of all-cause HRU in patients with IUI vs. without IUI during the 1-year follow-up period; comparative analysis: all abdominopelvic surgeries, GI surgeries, and GYN surgeries Parameter All abdominopelvic surgeries Difference a Patients with IUI ( N  = 216) Patients without IUI ( N  = 216) Estimate b (95% CI) P value Cohen’s d effect size All abdominopelvic surgeries, N  = 216 matched pairs Inpatient stay during index hospital case period c  Any inpatient admission, n (%) 208 (96.3) 201 (93.1) NR NR NR  Length of inpatient stay (days), d,e mean (SD) 19.1 (20.9) 9.4 (9.5) 9.6 (6.7–12.5) < 0.001 0.581  Any ICU stay, n (%) 30 (13.9) 11 (5.1) 2.7 (1.4–5.3) 0.002 NR Inpatient stay after index hospital case period  Any inpatient admission, n (%) 176 (81.5) 89 (41.2) 2.0 (1.7–2.3) < 0.001 NR  Length of inpatient stay (days), d mean (SD) 20.1 (21.0) 10.0 (24.8) 10.1 (5.7–14.5) < 0.001 0.439  Total number of inpatient admissions, d mean (SD) 2.5 (2.4) 1.0 (1.9) 1.5 (1.1–1.9) < 0.001 0.681 Outpatient visits  Any specialist visit, n (%) 197 (91.2) 203 (94.0) 1.0 (0.9–1.0) 0.270 NR  Any hospital visit, n (%) 92 (42.6) 34 (15.7) 2.7 (1.9–3.8) < 0.001 NR  Any emergency visit, n (%) 86 (39.8) 45 (20.8) 1.9 (1.4–2.6) < 0.001 NR  Any medication prescribed, n (%) 205 (94.9) 202 (93.5) 1.0 (1.0–1.1) 0.536 NR  Any medical devices utilized, n (%) 82 (38.0) 49 (22.7) 1.7 (1.2–2.3) 0.001 NR Parameter GI surgeries Difference a Patients with IUI ( n  = 94) Patients without IUI ( n  = 94) Estimate b (95%Cl) P value Cohen’s d effect size GI surgeries, n  = 94 matched pairs Inpatient stay during the index hospital case period  Any inpatient admission, n (%) 92 (97.9) 88 (93.6) NR NR NR  Length of inpatient stay (days), d,e mean (SD) 24.8 (23.0) 10.2 (8.3) 14.6 (9.7–19.5) < 0.001 0.826  Any ICU stay, n (%) 16 (17.0) 8 (8.5) 2.0 (0.9–4.4) 0.080 NR Inpatient stay after the index hospital case period  Any inpatient admission 74 (78.7) 46 (48.9) 1.6 (1.3–2.0) < 0.001 NR  Length of inpatient stay (days), d mean (SD) 21 (23.9) 11.6 (21.9) 9.4 (2.2–16.6) 0.011 0.405  Total number of inpatient admissions, d mean (SD) 2.4 (2.7) 1.3 (1.9) 1.1 (0.4–1.8) 0.002 0.487 Outpatient visits  Any specialist visit, n (%) 83 (88.3) 86 (91.5) 1.0 (0.9–1.1) 0.468 NR  Any hospital visit, n (%) 35 (37.2) 14 (14.9) 2.5 (1.4–4.3) < 0.001 NR  Any emergency visit, n (%) 36 (38.3) 19 (20.2) 1.9 (1.2–3.1) 0.006 NR  Any medication prescribed, n (%) 87 (92.6) 89 (94.7) 1.0 (0.9–1.1) 0.551 NR  Any medical devices utilized, n (%) 41 (43.6) 28 (29.8) 1.5 (1.0–2.2) 0.049 NR Parameter GYN surgeries Difference a Patients with IUI ( n  = 92) Patients without IUI ( n  = 92) Estimate b (95% CI) P value Cohen’s d effect size GYN procedures, n  = 92 Inpatient stay during index hospital case period  Any inpatient admission, n (%) 86 (93.5) 83 (90.2) NR NR NR  Length of inpatient stay (days), d,e mean (SD) 10.9 (8.9) 6.0 (5.5) 4.8 (3.1–6.6) < 0.001 0.630  Any ICU stay, n (%) 9 (9.8) NR NR NR NR Inpatient stay after index hospital case period  Any inpatient admission 78 (84.8) 23 (25.0) 3.4 (2.4–4.9) < 0.001 NR  Length of inpatient stay (days), d mean (SD) 18.4 (16.8) 3.0 (9.9) 15.4 (12.0–18.8) < 0.001 1.075  Total number of inpatient admissions, d mean (SD) 2.5 (2.1) 0.4 (0.8) 2.1 (1.7–2.5) < 0.001 1.255 Outpatient visits  Any specialist visit, n (%) 89 (96.7) 91 (98.9) 1.0 (0.9–1.0) 0.312 NR  Any hospital visit, n (%) 45 (48.9) 14 (15.2) 3.2 (1.9–5.4) < 0.001 NR  Any emergency visit, n (%) 40 (43.5) 17 (18.5) 2.4 (1.4–3.8) < 0.001 NR  Any medication prescribed, n (%) 91 (98.9) 87 (94.6) 1.0 (1.0–1.1) 0.097 NR  Any medical devices utilized, n (%) 28 (30.4) 9 (9.8) 3.1 (1.6–6.2) < 0.001 NR CI confidence interval, GI gastrointestinal, GYN gynecological, HRU healthcare resource utilization, ICU intensive care unit, IUI intraoperative ureteral injury, NR not reported, SD standard deviation a Relative risks (95% CI) and P  values were calculated for binary outcomes; for continuous outcomes, mean differences (95% CI), P  values, and standardized effect sizes (Cohen’s d ) were reported b For continuous outcomes (length of inpatient stay [days] and total number of admissions), the difference estimate was the mean of the paired differences. For binary outcomes, the difference estimate was calculated as the relative risk with 95% CI. Statistical significance for continuous outcomes was assessed using paired t  tests, while McNemar’s test was used for binary outcomes c Patients with invalid discharge dates were not included in this analysis d Per patient e Length of inpatient stay was described only in patients who had accurate end dates for their hospital case period (208 patients with IUI and 201 patients without IUI) Propensity score-matched analysis of all-cause HRU in patients with IUI vs. without IUI during the 1-year follow-up period; comparative analysis: all abdominopelvic surgeries, GI surgeries, and GYN surgeries CI confidence interval, GI gastrointestinal, GYN gynecological, HRU healthcare resource utilization, ICU intensive care unit, IUI intraoperative ureteral injury, NR not reported, SD standard deviation a Relative risks (95% CI) and P  values were calculated for binary outcomes; for continuous outcomes, mean differences (95% CI), P  values, and standardized effect sizes (Cohen’s d ) were reported b For continuous outcomes (length of inpatient stay [days] and total number of admissions), the difference estimate was the mean of the paired differences. For binary outcomes, the difference estimate was calculated as the relative risk with 95% CI. Statistical significance for continuous outcomes was assessed using paired t  tests, while McNemar’s test was used for binary outcomes c Patients with invalid discharge dates were not included in this analysis d Per patient e Length of inpatient stay was described only in patients who had accurate end dates for their hospital case period (208 patients with IUI and 201 patients without IUI) Overall, a greater proportion of patients with IUI had an emergency visit during the follow-up period ( P  < 0.001) compared with patients without IUI. In addition, medical device use was significantly higher in patients with IUI compared with patients without (38.2% vs. 22.7%; P  = 0.001). There were no statistically significant differences between patients with and without IUI regarding having an outpatient specialist visit (91.2% vs. 94.0%; P  = 0.270) or having any medication prescribed (94.9% vs. 93.5%; P  = 0.536). For patients who underwent a GI surgery (97.9%), nearly all patients with IUI required inpatient care during the 1-year follow-up period, vs. 93.6% of those without IUI. Patients with IUI stayed in the hospital an average of 14.6 days longer than those without IUI ( P  < 0.001). After discharge, a higher proportion of patients with IUI had inpatient readmission compared with patients without IUI (78.7% vs. 48.9%, P  < 0.001). A greater proportion of patients with IUI utilized outpatient services compared with patients without IUIs, including having any hospital visit (37.2% vs. 14.9%; P  < 0.001) and any emergency care visit (38.3% vs. 20.2%; P  = 0.006) during the follow-up period. A similar pattern of results was observed among the 92 matched pairs of patients with a GYN surgery. During the initial surgery, more patients with IUI needed inpatient care compared with patients without IUI (93.5% vs. 90.2%, respectively). Patients with IUI stayed in the hospital an average of 4.8 days longer than those without IUI ( P  < 0.001). After discharge, a significantly higher proportion of patients with IUI had a hospital stay compared with patients without IUI (84.8% vs. 25.0%; P  < 0.001). Use of any outpatient services was also significantly higher among patients with IUI vs. without IUI, including any outpatient hospital visit (48.9% vs. 15.2%, P  < 0.001) and any emergency care visit (43.5% vs. 18.5%; P  < 0.001). For all abdominopelvic surgeries, the total all-cause 1-year healthcare costs were significantly higher among patients with IUI (mean €67,041 [SD 103,807]) vs. without IUI (mean €38,177 [SD 81,389]); mean difference €28,864; P  = 0.001 (Table  3 ; Fig.  3 ). The cost of inpatient care was significantly higher among patients with IUI than among those without (€26,046 vs. €13,600; P  < 0.001). The mean outpatient cost for patients with IUI was €40,848, compared with €24,537 for patients without IUI, a difference that approached statistical significance ( P  = 0.053). Medication costs, although substantial, showed wide variability. For all abdominopelvic surgeries, as well as GI and GYN surgeries, inpatient hospital costs were significantly higher among patients with IUI compared with those without IUI (all P  values < 0.01). Outpatient costs were higher among patients with IUIs as well, although the only difference that was statistically significant was observed among patients who underwent GYN surgeries ( P  = 0.001). For patients with IUIs who had GYN and GI surgeries, mean (SD) total costs were €59,593 (76,628) and €69,366 (130,374), respectively. Table 3 All-cause healthcare costs in propensity score-matched patients with and without IUI during the 1-year follow-up period; comparative analysis Costs (€), mean (SD) Patients with IUI, mean (SD) Patients without IUI, mean (SD) Mean difference, 95% CI P value All abdominopelvic surgeries, N  = 216  Overall healthcare costs (inpatient and outpatient) 67,041 (103,807) 38,177 (81,389) 28,864 (11,647 to 46,081) 0.001 Inpatient hospital costs a 26,046 (23,861) 13,600 (15,412) 12,445 (8842 to 16,048) < 0.001 Outpatient overall costs b 40,846 (96,070) 24,537 (77,168) 16,309 (− 232 to 32,850) 0.053 Outpatient medication costs c 31,004 (84,416) 17,036 (68,028) 13,969 (− 820 to 28,757) 0.064 GI procedures, n  = 94  Overall healthcare costs (inpatient and outpatient) 59,593 (76,628) 44,536 (87,114) 15,057 (− 8694 to 38,808) 0.211 Inpatient hospital costs a 28,491 (19,729) 14,857 (12,689) 13,634 (8877 to 18,390) < 0.001 Outpatient overall costs b 31,102 (67,980) 29,628 (83,837) 1474 (− 21,323 to 24,271) 0.898 Outpatient medication costs c 20,863 (58,573) 20,403 (76,084) 461 (− 19,805 to 20,727) 0.964 GYN procedures, n  = 92  Overall healthcare costs (inpatient and outpatient) 69,366 (130,374) 15,010 (26,507) 54,356 (28,067 to 80,645) < 0.001 Inpatient hospital costs a 18,731 (13,166) 7184 (9606) 11,548 (8462 to 14,633) < 0.001 Outpatient overall costs b 50,635 (125,015) 7826 (23,380) 42,808 (17,031 to 68,585) 0.001 Outpatient medication costs c 41,835 (113,874) 4847 (21,644) 36,988 (13,434 to 60,541) 0.002 Costs are shown in euros GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury, SD standard deviation a Inpatient hospital costs, the sum of all costs of inpatient hospitalizations with ≥ 1 overnight stay, were derived from the sum of all costs attributed to the respective inpatient case in the observational period of “hospital treatment” b Outpatient overall costs, the sum of all costs for 4 outpatient services/settings: outpatient hospital, outpatient GP, outpatient specialist, and outpatient emergency visit costs; sum of all case-associated costs plus all case points multiplied by the respective “reference point value” per year for each case c Outpatient medication costs, the sum of all costs for outpatient dispensed drugs, identified by ABDA as medication, including costs of special codes (“special pharmaceutical registration number”) for individual preparation Fig. 3 Differences in all-cause mean healthcare costs among matched pairs of patients with and without IUI during the 1-year follow-up period; comparative analysis: all abdominopelvic surgeries, GI surgeries, and GYN surgeries. GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury All-cause healthcare costs in propensity score-matched patients with and without IUI during the 1-year follow-up period; comparative analysis Costs are shown in euros GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury, SD standard deviation a Inpatient hospital costs, the sum of all costs of inpatient hospitalizations with ≥ 1 overnight stay, were derived from the sum of all costs attributed to the respective inpatient case in the observational period of “hospital treatment” b Outpatient overall costs, the sum of all costs for 4 outpatient services/settings: outpatient hospital, outpatient GP, outpatient specialist, and outpatient emergency visit costs; sum of all case-associated costs plus all case points multiplied by the respective “reference point value” per year for each case c Outpatient medication costs, the sum of all costs for outpatient dispensed drugs, identified by ABDA as medication, including costs of special codes (“special pharmaceutical registration number”) for individual preparation Differences in all-cause mean healthcare costs among matched pairs of patients with and without IUI during the 1-year follow-up period; comparative analysis: all abdominopelvic surgeries, GI surgeries, and GYN surgeries. GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury GYN procedures had a higher incidence proportion of IUIs (164.7 per 100,000 surgeries) than GI procedures (102.1 per 100,000 surgeries), highlighting the elevated risk associated with gynecological surgeries. The cost differences in GYN surgeries were particularly notable, with patients with IUI incurring substantially higher overall healthcare expenses (€69,366 vs. €15,010 for patients without IUI; P  < 0.001). Medication costs were the primary driver of these differences, with patients with IUI spending €41,835 compared with €4847 for patients without IUI ( P  = 0.002). GYN procedures exhibited a higher incidence proportion of IUIs (164.7 per 100,000 surgeries) compared with GI procedures (102.1 per 100,000 surgeries), highlighting the elevated risk associated with gynecological surgeries. In GYN procedures, patients with IUI showed considerably higher HRU compared with patients without IUIs. A majority of patients with IUIs (84.8%) needed additional hospital stays after surgery vs. 25.0% of patients without IUI. Furthermore, patients with IUI also faced higher outpatient costs, especially for medical devices and emergency visits.

This analysis of the WIG2 claims database provides valuable real-world evidence on the incidence and healthcare burden of IUIs in Germany. Comparative analyses revealed that patients who experienced an IUI incurred substantially higher all-cause HRU and costs over the year following surgery compared with patients without an IUI. These findings underscore the clinical and economic impact of IUIs and highlight the importance of continued efforts to prevent, promptly identify, and optimally manage these injuries in surgical practice.

Although IUIs are relatively rare, they represent a serious and potentially life-threatening complication of abdominopelvic surgeries. IUIs are linked to elevated risks of postoperative complications, prolonged recovery periods, and increased mortality rates. The severity of these outcomes highlights the critical importance of timely detection and effective management to mitigate the long-term effects on patient health. This study of the WIG2 claims database in a German population revealed that the incidence of IUIs varied depending on the type of surgery performed, with certain procedures such as proctectomy and complex hysterectomies showing higher incidence of IUIs. Comparative analyses showed that patients who experienced IUI had longer hospital stays compared with those without IUI, suggesting increased clinical complexity of these cases. Moreover, these injuries were associated with higher healthcare costs, showing the considerable economic impact of IUIs on the healthcare system. These findings emphasize the substantial burden that IUI imposes on both patients and healthcare systems, underscoring an urgent need for enhanced preventive measures and surgical care strategies. Improved intraoperative techniques and tools, better training, and increased awareness among surgical teams could reduce the incidence of IUIs. Additionally, early recognition and prompt intervention are essential to help minimize complications, reduce hospital stays, and contain costs, ultimately improving patient outcomes and healthcare efficiency [ 3 ]. While direct inspection of the ureters with ureteroscopy is currently the most sensitive method for detecting IUI, intravenous dyes and imaging agents may also offer additional value to check for leakage along the ureter. However, limitations in the pharmacokinetic and structural properties of available dyes may lead to adverse effects, such as arrhythmias, coronary vasoconstriction, and hemolytic anemia. Therefore, there is a need for novel, minimally invasive intraoperative imaging agents to improve visualization of the ureter during surgical procedures [ 16 ]. A key strength of this study was its use of a large, real-world claims database, which enabled evaluation of IUI incidence, HRU, and healthcare costs in routine care. To improve case identification beyond reliance on a single diagnosis code, a refined three-tier classification system for IUI was developed in collaboration with clinical experts. This included three distinct definitions for IUIs: definite IUI (identified solely by ICD-10-GM code S37.1); moderate (probable) IUI (combining ICD-10-GM and OPS codes to capture likely but uncoded injuries); and broad (probable) IUI (a more inclusive definition used in sensitivity analyses). This approach enhanced diagnostic precision and mitigated potential bias from under- or over-estimation. Several limitations of this study should be considered. As with all secondary analyses of claims data, there was potential for coding misclassification, particularly with ICD-10-GM diagnoses, and HRU and healthcare costs were interpreted as all-cause rather than directly attributed to IUIs. The algorithm used to identify IUI cases, though rigorously developed and validated through expert input, blinded case review, and benchmarking against published data, may still have had limitations in sensitivity or specificity. Additionally, although a two-stage matching approach was employed—first exactly matching on surgical procedure and setting to create procedurally homogeneous strata, then applying PSM within each stratum to balance clinical and sociodemographic characteristics—residual confounding cannot be fully excluded. The heterogeneity of abdominopelvic procedures included in this study means that even within matched strata, unmeasured procedure-specific factors (e.g., anatomical complexity, surgeon volume, or intraoperative technique) not captured in claims data may have differentially influenced outcomes between patients with and without IUI. Future studies with larger IUI cohorts should consider incorporating more granular procedure-level covariates to further isolate the independent contribution of IUI to HRU and costs. Despite these considerations, the combination of robust methodology and real-world data contributes valuable insights into the burden of IUIs in a German population.

Intraoperative ureteral injury (IUI) is a rare but serious complication of abdominopelvic surgery [ 1 – 3 ] and can lead to significant adverse outcomes, including death [ 4 – 6 ]. For instance, in 2014, Halabi and colleagues, using a large population-based analysis to compare cases with vs. without documented IUI, found that IUI substantially increased the risk of sepsis, urinary fistula, and acute renal insufficiency, as well as increased the risk of death [ 5 ]. Gynecological (GYN), obstetric, colorectal, urological, and vascular surgical procedures are most frequently associated with IUI [ 4 – 6 ]. However, these injuries often go undetected during the initial procedure and may only be diagnosed days or weeks later [ 7 ]. Delayed recognition of IUI is associated not only with significantly higher morbidity but also with increased numbers of secondary procedures, hospital admissions, and prolonged hospital stays [ 7 ]. Findings from a US claims database showed that patients with IUI experienced increased healthcare resource utilization (HRU), including a 15.3% higher 30-day hospital readmission rate compared to those without IUI. The majority of the associated costs were primarily attributable to inpatient services [ 8 ]. Diagnostic cystoscopy with retrograde pyelography is the gold standard for detecting IUI, but its routine use varies across institutions and countries [ 1 , 3 , 9 ]. Emerging techniques such as fluorescence-guided imaging offer potential advantages in real-time ureteral visualization; however, these methods have not yet been widely adopted [ 1 , 10 , 11 ]. Despite the clinical significance of IUI, real-world data on its incidence and economic burden are scarce, posing a challenge to the development of targeted prevention strategies and cost-effective management. To help close this gap, this study leverages a comprehensive German claims database to investigate the incidence, associated HRU, and related costs among patients with IUI compared with patients without IUI. By offering real-world insights, these findings aim to inform clinical decision-making and support initiatives to mitigate both the health and economic consequences of IUI.

This descriptive, comparative, longitudinal, observational cohort study used anonymized claims data from the German Wissenschaftliches Institut für Gesundheitsökonomie und Gesundheitssystemforschung (WIG2) benchmark database (Fig.  1 ). Adults (aged ≥ 18 years) who underwent abdominopelvic surgery with a risk of IUI between 1 January 2014 and 31 December 2021 were included, while patients with prior IUI diagnosis, ureter reconstruction, or trauma to the abdominopelvic area within 6 months before surgery, or ureteral calculi treatment on the surgery date were excluded. The index date was defined as the date of the surgery with a risk of IUI. The follow-up period was defined as the index date and 1 year after the index date. The follow-up period was divided into early (0–30 days), middle (31–90 days), and late (91–365 days) phases. A lookback period of ≤ 2 years before the index date was used to assess patient comorbidities and risk factors. All-cause HRU and healthcare costs were assessed during the 1-year period following the index date. The study design accounted for stratification by surgical type, procedure, approach, setting, and admission reason to ensure a thorough analysis. Additionally, the study included a validation cohort of patients undergoing low-risk surgeries to develop and validate an algorithm for identifying probable IUI cases. Fig. 1 Study design. GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury Study design. GI gastrointestinal, GYN gynecological, IUI intraoperative ureteral injury This study used the WIG2 database, an anonymized claims dataset covering approximately 4.5 million individuals insured by German statutory health insurers (SHIs). The dataset is considered representative of the broader population covered by SHI in terms of age, sex, and morbidity [ 12 – 14 ] and included data from January 2014 to 31 December 2021. Available information included demographic data, full billing records for inpatient and outpatient care, and reimbursed treatments. Diagnosis codes from outpatient care were provided quarterly, and dates were available for outpatient procedures and drug dispensations. All data in the WIG2 benchmark database were anonymized to comply with German privacy regulations. Their use for scientific purposes was in conformity with German law; and no additional permissions were required from an institutional review board (IRB) or an independent ethics committee (IEC). This study was conducted in compliance with all applicable German, national, and European Union requirements. All applicable data protection and privacy regulations were observed in collecting, forwarding, processing, and storing participants’ data. Patients aged 18 years or older who underwent abdominopelvic surgery were identified using operation and procedure (Operationen- und Prozedurenschlüssel [OPS]) codes informed by published literature, internal analyses, and clinical expertise. The primary objective of this study was to estimate the IUI incidence proportion in the 90-day period following index abdominopelvic surgeries, both overall and stratified by surgical type (gastrointestinal [GI], GYN, and “other” abdominopelvic surgeries [i.e., aortic procedure, aortic procedure including infrarenal, iliac lymph system procedure, other lymph system procedure, para-aortic and iliac lymph system procedure, para-aortic lymph system procedure, retroperitoneum procedure, testicle procedures, and unclear internal procedure]) and by specific surgical procedures within surgical type, such as colectomy and hysterectomy. Select secondary objectives included descriptive statistics for patient and surgical characteristics, including the overall population and subgroups by IUI status and surgical type. Additionally, comparative analyses of HRU between patients with and without IUI were performed, overall and by surgical type. Before completing the comparative analyses, we validated the provisional list of codes to identify IUIs using a moderate (probable) definition and finalized the method used to identify patients with IUI. These definitions are further detailed in the next section. Comparative analyses estimated the differences in selected all-cause HRU and healthcare costs between patients with and without IUI overall and by surgical type. IUI cases were identified using the International Classification of Diseases, Tenth Revision, German Modification (ICD-10-GM) diagnosis codes or a combination of selected ICD-10-GM codes and OPS codes. To account for the likely underreporting of IUI using ICD-10-GM code S37.1 (injury to ureter), which we refer to here as the “strict” definition, an algorithm was developed to examine two additional definitions of IUI: moderate and broad (see Fig.  S1 for an overview and Table  S1 for the list of codes used for each definition). The moderate definition included the strict definition as well as diagnosis codes indicating symptoms of IUI in combination with ureteral reconstruction. The broad definition included the codes for the strict and moderate definitions, as well as additional diagnosis and procedure codes indicating symptoms and procedures that may be associated with IUI. These combinations were first verified in the exploratory analysis conducted before the HRU and cost analyses. Please see Supplementary Information for detailed methodological considerations for the IUI definition algorithm. Patient characteristics were evaluated for the 2 years before surgery. Patient variables included age, sex, Charlson Comorbidity Index (CCI), and cancer status. Surgical variables included type (GI, GYN, or other abdominopelvic procedures), procedure (e.g., colectomy, hysterectomy), setting (inpatient vs. outpatient), and approach (open [laparotomy], minimally invasive [laparoscopy], or unknown). Variables were stratified by IUI status, surgical type, and procedure. The index hospital stay was defined as the inpatient admission date through the high trim point (6 to 10 days for less complex surgeries and > 30 days for extremely complex cases), as defined in the German Diagnosis-Related Groups (DRG) System. The high trim point represents the upper statistical boundary in days for the normal length of stay of patients assigned to a given DRG. Please see Supplementary Information for detailed methods related to index hospitalization. HRU outcomes were assessed during and after the index hospital stay, including duration of hospitalization; any inpatient admissions; any emergency visits; any specialist consultations; any prescribed medications; and any medical device use. Intensive care unit use was assessed only during the index hospital stay. Healthcare cost outcomes included inpatient hospital costs, outpatient overall costs (including outpatient hospital, outpatient general practitioner, outpatient specialist, and outpatient emergency visits), total healthcare costs (sum of inpatient hospital costs and outpatient overall costs), and outpatient medication costs. All data in the WIG2 benchmark database were anonymized to comply with German privacy regulations, their use for scientific purposes was in conformity with German law, and no additional permissions were needed from an institutional review board or an independent ethics committee. During the study, Tobias Heidler and Nils Kossack, employees of the WIG2 Institute, had permission to access the database. This study was conducted in compliance with all applicable requirements in Germany for ensuring the rights of participants in non-interventional studies. The study was also conducted in compliance with national and European Union requirements for ensuring the rights of participants in non-interventional studies. All applicable data protection and privacy regulations were observed in collecting, forwarding, processing, and storing participants’ data. Continuous variables (e.g., age, CCI) were summarized using means with standard deviations, medians with interquartile ranges (IQR), and ranges. Categorical variables were summarized using frequencies and proportions. IUI incidence proportions were calculated for all abdominopelvic surgeries and stratified by surgical type and setting. The numerator was the number of patients with IUI, and for stratified analyses, the denominator was the total number of patients who underwent the corresponding subtype of abdominopelvic surgeries. Incidence proportions were standardized per 100,000 patients, with 95% confidence intervals (CIs) adjusted for the Agresti–Coull method [ 15 ]. Comparative analyses evaluated all-cause HRU and healthcare costs between patients with and without IUI during the 1-year follow-up period. Patients without IUI were matched 1:1 to patients with IUI using exact matching by surgical setting (inpatient or outpatient) and surgical type prior to PSM. To account for potential differences in other clinical and sociodemographic characteristics between those with and without IUI, 1:1 PSM was employed. Within each matching stratum, propensity scores were estimated using logistic regression, adjusting for age, sex, obesity (yes/no), CCI score, current cancer diagnosis, prior history of endometriosis, number of prior abdominopelvic surgeries, prior history of diverticular disease, prior history of abdominopelvic perforation, and prior history of IUI. To achieve successful PSM (i.e., < 10% standardized mean difference), a reduction in the number of categories for prior abdominopelvic surgeries was required in some strata to avoid overfitting in the logistic regression. Exact matching between surgical type, surgical procedure, and surgical setting was performed. A nearest-neighbor method without replacement was used, with a caliper of 0.2 standard deviations. Covariate balance was assessed using standardized mean differences, with values < 10% indicating acceptable balance. If balance was not achieved, models were refined by adjusting variable forms or including interaction terms. Procedures with < 5 cases were censored to ensure accurate incidence estimates and to maintain privacy, the latter per WIG2 censoring procedures. To ensure reliable comparisons, comparative analyses were conducted only in strata with ≥ 46 matched pairs, which provided 90% power to detect a moderate effect size at the 0.05 significance level. Following PSM, HRU, and healthcare cost outcomes were compared among matched pairs with and without IUI. For outcomes that were continuous variables, the mean difference between matched pairs, 95% CIs, effect size (Cohen’s d ), and P  values from the paired t  test are reported. Effect sizes were reported for continuous variables only. Analyses were performed for all abdominopelvic surgeries and were stratified by surgical type without adjustment for multiple comparisons.

Below is the link to the electronic supplementary material. Supplementary file1 (PDF 246 kb) Supplementary file1 (PDF 246 kb)