Abstract
Objectives:
To describe complications at the time of surgery, 90-day readmission and 1-year reoperation rates after minimally-invasive pelvic organ prolapse (POP) in women ≥ 65 years of age in the United States (US) using Medicare 5% Limited Data Set (LDS) Files.
Methods:
Medicare is a federally funded insurance program in the US for individual 65 and older. Currently, 98% of individuals over the age of 65 in the US are covered by Medicare. We identified women undergoing minimally invasive POP surgery, defined as laparoscopic or vaginal surgery, in the inpatient and outpatient settings between 2011–2017. Patient and surgical characteristics as well as adverse events were abstracted. We used logistic regression for complications at index surgery, and cox proportional hazards regression models for time to readmission and time to reoperations.
Results:
11,779 women met inclusion criteria. The mean age was 72 years (SD±8); the majority were White (91%). The majority of procedures were vaginal (76%) and did not include hysterectomy (68%). The rate of complications was 12%; vaginal hysterectomy (aOR 2.4, 95%CI 2.2–2.7) was the factor most strongly associated with increased odds of complications. The 90-day readmission rate was 7.3%. The most common reason for readmission was infection (2.0%), three quarters of which urinary tract infections. Medicaid eligibility (aHR 1.5, 95%CI 1.3–1.8) and concurrent sling procedures (aHR 1.2, 95%CI 1.04–1.4) were associated with a higher risk of 90-day readmission. The 1-year reoperation rate was 4.5%. The most common type of reoperation was a sling procedure (1.8%). Obliterative POP surgery (aHR 0.6, 95%CI 0.4–0.9) was associated with a lower risk of reoperation than other types of surgery.
Conclusion:
US women 65 years and older who are also eligible to receive Medicaid are at higher risk of 90-day readmission following minimally-invasive surgery for POP with the most common reason for readmission being UTI.
Keywords: minimally invasive surgery, pelvic organ prolapse, readmission, reoperation, complications
Brief summary:
US women 65 years and older who are also eligible to receive Medicaid are at higher risk of 90-day readmission following minimally invasive surgery for POP with the most common reason for readmission being UTI.
Introduction:
The lifetime risk of surgery for pelvic floor disorders is estimated to be 20%[1], and as the population ages, a large proportion of American women will need surgical management for pelvic organ prolapse (POP).[2] Increasing age [3–5] and frailty [6] have been independently associated with adverse postoperative outcomes. Prior studies have identified a number of patient factors associated with adverse outcomes following POP surgery such as smoking status [7], body mass index[5,8], medical comorbidities[9] and history of a prior POP surgery[10]. Surgical characteristics such as use of mesh for POP repair [11–13], concomitant surgery for stress incontinence [11,14], route and type of POP surgery.[10,15,16] and surgeon volume [5,17] [18,19] have also been identified as potentially modifiable risk factors that impact the frequency and types of adverse events. Recognizing factors associated with complications, readmissions and reoperations following POP surgery can improve outcomes, uncover disparities and variations in care, and identify mechanisms to help mitigate the cost burden associated with adverse outcomes in this population.
To date, literature on adverse outcomes following POP surgery has largely been limited to the acute postoperative period, typically less than 30 days after surgery. Use of national databases, such as the Medicare 5% Limited Data Set (LDS) database, allows for prolonged longitudinal follow up of a large, diverse and nationally representative sample of older patients after POP surgery. Medicare is a federally funded insurance program in the United States (US) for individual 65 and older. As of 2019, there are more than 54 million older adults age 65 and older living in the US, 98% of whom are covered by Medicare. By 2050, that number is expected to reach nearly 90 million. [20] As this segment of the population grows, it is important to understand factors associated with adverse outcomes following elective surgery for POP. Thus, the primary aim of this study was to describe adverse outcomes defined as complications at the time of index surgery, 90-day readmissions and 1-year reoperations following POP surgery in women over 65 between 2011 and 2017. Secondary aims were to identify patient and surgical factors associated with adverse outcomes following POP surgery and to describe costs associated with adverse outcomes following POP surgery.
Materials and Methods:
For this study, we identified women undergoing minimally invasive surgery, i.e. laparoscopic or vaginal, for POP using the Medicare 5% LDS database. The Medicare 5% LDS reflects a random sampling of 5% of all Medicare claims across the nation. The Medicare database is unique in that it is the largest collection of longitudinal hospital care data in the US. Given that it captures over 98% of Americans aged 65 and over, strengths of the database include its generalizability to the older US population and its potential for well-powered analyses. It provides data on claims and costs in many different settings including inpatient, outpatient, skilled-nursing facilities and hospice settings. Also, once patients enroll in Medicare, withdrawal from the program is very rare; thus, health information is typically collected up to the death of a Medicare beneficiary. Of note, information regarding patient behavior such as diet, exercise and substance use as well as results from diagnostic tests are not available in Medicare datasets.
Regarding inclusion criteria, we identified women undergoing minimally invasive surgery, i.e. laparoscopic or vaginal, for POP in the inpatient and outpatient settings between 2011–2017 using Current Procedural Terminology (CPT) codes. As open abdominal surgery is an independent risk factor for complications following benign gynecologic surgery [21] [22], we chose to limit our study to minimally invasive procedures. Of note, open abdominal sacrocolpopexy, which was excluded from this cohort, represented a very small percentage of the cohort (<1%) who underwent any surgery for prolapse during our study period. The following procedures were included: Laparoscopic colpopexy, intraperitoneal colpopexy, extraperitoneal colpopexy, colpoclesis, vaginectomy, anterior repair, posterior repair, and combined anterior-posterior repair (CPT codes: 57425, 57289, 57283, 57120, 57110, 57240, 57250, 57260, 57265, 57268, 57282, 57267). We included patients who underwent surgery with a gynecologist or urologist. We excluded patients who were undergoing concomitant surgery for oncologic diagnoses and those undergoing surgery performed by another specialty operating within the pelvis or abdomen. Additional covariates of interest were collected: age, race/ethnicity, Medicaid status, geographical region (West, Midwest, South, Northeast) and Charlson Comorbitidy Score[23]. Patients are eligible for Medicaid if they have an income under 133% of the federal poverty level [24], and thus we chose to use Medicaid status as a proxy for low socioeconomic status. In terms of procedure characteristics, covariates of interest included concurrent sling, use of vaginal mesh, type of hysterectomy (none, laparoscopic and vaginal), and type of POP surgery, which was further categorized into laparoscopic, vaginal and obliterative procedures. Laparoscopic procedures included laparoscopic sacrocolpopexy; vaginal procedures included intraperitoneal colpopexy, extraperitoneal colpopexy, anterior repair, posterior repair and anterio-posterior repair; and obliterative procedures included colpoclesis and vaginectomy. This study was deemed exempt by the Institutional Review Board.
We estimated the incidence of complications, 90-day readmission, and 1-year reoperation in this study population. We identified complications and reasons for readmission using ICD codes and then categorized them by organ system. Reoperations included surgery for POP, stress incontinence, and conditions directly related to index surgery such as removal or revision of sling for incontinence, revision of graft laparoscopic and vaginal approaches, ureteral stent placement, vesicovaginal fistula repair, urethrovaginal fistula repair, ureteroneocystotomy, and ureteroureterostomy (CPT codes: 57425, 57289, 57283, 57120, 57110, 57240, 57250, 57260, 57265, 57268, 57282, 57267, 57288, 57287, 52332, 51900, 51999, 57320, 57330, 51925, 57310, 57311, 50780, 50760). We chose to define our readmission period to 90-days to capture admissions falling outside the immediate postoperative period. Also, 90-days reflect the global period following major surgery as determined by Medicare. The global period is the duration of time that a physician may not bill for related outpatient visits, admissions, procedures or surgeries following the index procedure. [25]
We described overall baseline characteristics of patients who underwent surgery for POP during the study period, and compared these characteristics by adverse outcome status. We compared patient groups using chi-square tests for categorical variables and Wilcoxon rank-sum tests for continuous variable. We estimated the cumulative incidence for readmissions over 90 days and reoperations over 365 days from the index surgery date with Kaplan Meier curves.To estimate the unadjusted and adjusted effects of candidate variables on adverse outcomes, we used univariate and multivariable regression models; specifically, logistic regression for complications at index surgery and cox proportional hazards regression models for time to readmission and time to reoperations. All models controlled for age, race, Medicaid eligibility, geographic region, type of index POP surgery, hysterectomy, vaginal mesh, sling and Charlson Comorbidity Score. We chose to control for these demographic and surgical factors based on existing literature that suggests an association with outcomes following gynecologic surgery. We adjusted cost variables across years (2011–2018) to the most recent year ($2018) using the medical care component of the Consumer Price Index. Also, we removed outliers by a 1% and 99% Winsorization of cost totals. We then summarized costs by groups of interest. We then tested for differences in costs between groups were assessed using Wilcoxon rank-sum as appropriate. R version 4.0.3 was used to perform statistical analysis. A two-sided p value <.05 was considered significant.
Results:
11,779 women met inclusion criteria, among whom the mean age was 72 ± 8 years; the majority were White (91%) and from the South (44%) (Table 1). The majority of procedures were performed vaginally (76%) and did not include hysterectomy (68%). During the index surgical encounter, 12% experienced a complication, the majority of which were associated with the genitourinary tract (6.8%). Medicaid eligibility differed significantly between those who experienced a complication and those who did not (15% vs 13%, p = 0.01). Regarding surgical characteristics, type of POP surgery (vaginal, laparoscopic, obliterative) (p<.001) and hysterectomy (p=0.001) was associated with complications. Undergoing concurrent sling was not associated with complications. After adjustment, Medicaid eligibility (aOR 1.2, 95%CI 1.04–1.3) was associated with a 20% higher odds of complications at index surgery (Table 2). In terms of surgical characteristics, vaginal hysterectomy (aOR 2.4, 95%CI 2.2–2.7) and obliterative surgery (aOR 1.5, 95%CI 1.3–1.8) were also associated with increased odds of complications. Women in the South (aOR 0.8, 95%CI 0.7–0.9) had a lower odds of complications compared to women in the Midwest.
Table 1.
Differences across cohort by adverse outcomes
| Overall cohort N = 11,779 |
No complication N = 10,372 |
Complication N = 1,407 |
p-value | No Reoperation N = 11,196 |
Reoperation N = 583 |
p | No Readmission n = 10,886 |
Readmission n = 893 |
p-value | |
|---|---|---|---|---|---|---|---|---|---|---|
| Age | 71.5 (8.2) | 71.3 (8.3) | 72.9 (8.7) | <0.001 | 71.5 (8.4) | 71.7 (7.7) | 0.49 | 71.4 (8.2) | 72.4 (10.1) | 0.001 |
| Race | 0.79 | 0.07 | <.001 | |||||||
| White | 10717 (91.0) | 9431 (90.9) | 1286 (91.4) | 10172 (90.8) | 545 (93.6) | 9898 (90.9) | 819 (91.7) | |||
| Black | 519 (4.4) | 458 (4.4) | 61 (4.3) | 501 (4.5) | 18 (3.1) | 484 (4.4) | 35 (3.9) | |||
| Other | 543 (4.6) | 483 (4.7) | 60 (4.3) | 524 (4.7) | 19 (3.3) | 504 (4.6) | 39 (4.4) | |||
| Region | <0.001 | 0.96 | 0.17 | |||||||
| Midwest | 2802 (23.8) | 2414 (23.3) | 388 (27.6) | 2668 (23.8) | 134 (23.0) | 2568 (23.6) | 234 (26.2) | |||
| Northeast | 1661 (14.1) | 1487 (14.3) | 174 (12.4) | 1580 (14.1) | 81 (13.9) | 1552 (14.3) | 109 (12.2) | |||
| South | 5137 (43.6) | 4599 (44.3) | 538 (38.2) | 4881 (43.6) | 256 (44.0) | 4754 (43.7) | 383 (42.9) | |||
| West | 2179 (18.5) | 1872 (18.0) | 307 (21.8) | 2068 (18.5) | 111 (19.1) | 2012 (18.5) | 167 (18.7) | |||
| Medicaid Eligible | 1512 (12.8) | 1301 (12.5) | 211 (15.0) | 0.01 | 1449 (12.9) | 63 (10.8) | 0.15 | 1355 (12.4) | 157 (17.6) | <0.001 |
| Prolapse procedure | <0.001 | 0.001 | 0.01 | |||||||
| Vaginal | 8713 (75.5) | 7753 (76.3) | 960 (69.5) | 8284 (75.4) | 429 (76.2) | 8075 (75.7) | 638 (72.7) | |||
| Laparoscopic | 1706 (14.8) | 1490 (14.7) | 216 (15.6) | 1605 (14.6) | 101 (17.9) | 1576 (14.8) | 130 (14.8) | |||
| Obliterative | 1126 (9.8) | 921 (9.1) | 205 (14.8) | 1093 (10.0) | 33 (5.9) | 1016 (9.5) | 110 (12.5) | |||
| Hysterectomy | 0.001 | 0.21 | 0.32 | |||||||
| None | 8030 (68.2) | 7125 (68.7) | 905 (64.3) | 7618 (68.0) | 412 (70.8) | 7402 (68.0) | 628 (70.3) | |||
| Vaginal | 2336 (19.8) | 2008 (19.4) | 328 (23.3) | 2237 (20.0) | 99 (17.0) | 2167 (19.9) | 169 (18.9) | |||
| Laparoscopic | 1413 (12.0) | 1239 (11.9) | 174 (12.4) | 1342 (12.0) | 71 (12.2) | 1317 (12.1) | 96 (10.8) | |||
| Vaginal mesh procedure | 2447 (20.8) | 2166 (20.9) | 281 (20.0) | 0.45 | 2296 (20.5) | 151 (25.9) | 0.002 | 2255 (20.7) | 192 (21.5) | 0.61 |
| Concurrent sling procedure | 4876 (41.4) | 4293 (41.4) | 583 (41.4) | 1.00 | 4627 (41.3) | 249 (42.8) | 0.51 | 4466 (41.0) | 410 (45.9) | 0.005 |
| Charlson Comorbidity Index Score | 1.0 [0.0, 2.0] | 1.0 [0.0, 2.0] | 1.0 [0.0, 3.0] | <0.001 | 1.0 [0.0, 2.0] | 1.0 [0.0, 2.0] | 0.43 | 1.0 [0.0, 2.0] | 1.0 [0.0, 3.0] | <0.001 |
Table 2.
Models for adverse outcomes
| Unadjusted Complications | Adjusted Complications | Unadjusted Readmission | Adjusted Readmission | Unadjusted Reoperation | Adjusted Reoperation | |
|---|---|---|---|---|---|---|
| OR (95%CI) | aOR (95%CI) | Hazard Ratio (95%CI) | Adjusted Hazard Ratio (95%CI) | Hazard Ratio (95%CI) | Adjusted Hazard Ratio (95%CI) | |
| Age | 1.02 (1.02 – 1.03) | 1.01 (1.01 – 1.02) | 1.01 (1.01 – 1.02) | 1.01 (1.00–1.02) | 1.00 (0.99 – 1.01) | 1.01 (0.99 – 1.02) |
| Race | ||||||
| White | -- | -- | -- | -- | -- | -- |
| Black | 0.98 (0.74 – 1.27) | 1.11 (0.91 – 1.35) | 0.88 (0.63 – 1.24) | 0.72 (0.50 – 1.02) | 0.68 (0.42 – 1.08) | 0.75 (0.46 – 1.20) |
| Other | 0.91 (0.69 – 1.19) | 1.01 (0.83 – 1.22) | 0.95 (0.69 – 1.31) | 0.78 (0.56 – 1.09) | 0.70 (0.44 – 1.10) | 0.77 (0.48 – 1.23) |
| Region | ||||||
| Midwest | -- | -- | -- | -- | -- | -- |
| Northeast | 0.73 (0.60 – 0.88) | 1.06 (0.93 – 1.20) | 0.78 (0.62 – 0.97) | 0.72 (0.57 – 0.91) | 1.01 (0.77 – 1.33) | 1.05 (0.79 – 1.38) |
| South | 0.73 (0.63 – 0.84) | 0.75 (0.68 – 0.83) | 0.89 (0.75 – 1.04) | 0.91 (0.77 – 1.08) | 1.03 (0.84 – 1.27) | 1.04 (0.84 – 1.29) |
| West | 1.02 (0.87 – 1.20) | 1.13 (1.01 – 1.29) | 0.91 (0.75 – 1.11) | 0.90 (0.74 – 1.11) | 1.05 (0.82 – 1.36) | 1.01 (0.78 – 1.31) |
| Medicaid Eligible | 1.23 (1.05 – 1.44) | 1.18 (1.04 – 1.34) | 1.48 (1.24 – 1.75) | 1.51 (1.25 – 1.83) | 0.83 (0.64 – 1.08) | 0.93 (0.70 – 1.31) |
| Prolapse procedure | ||||||
| Vaginal | -- | -- | -- | -- | -- | -- |
| Laparoscopic | 1.17 (0.99 – 1.37) | 1.20 (1.05 – 1.37) | 1.05 (0.87 – 1.27) | 1.17 (0.95 – 1.45) | 1.23 (0.99 – 1.53) | 1.37 (1.07 – 1.75) |
| Obliterative | 1.80 (0.93 – 1.31) | 1.27 (1.10 – 1.47) | 1.35 (1.10 – 1.65) | 1.16 (0.93 – 1.45) | 0.59 (0.42 – 0.84) | 0.59 (0.41 – 0.86) |
| Hysterectomy | ||||||
| None | -- | -- | -- | -- | -- | -- |
| Vaginal | 1.29 (1.12 – 1.47) | 2.44 (2.21 – 2.71) | 0.93 (0.78 – 1.10) | 1.04 (0.87 – 1.25) | 0.83 (0.67 – 1.04) | 0.91 (0.73 – 1.15) |
| Laparoscopic | 1.11 (0.93 – 1.31) | 1.16 (1.01 – 1.34) | 0.87 (0.70 – 1.08) | 0.93 (0.73 – 1.18) | 0.98 (.076 – 1.26) | 0.92 (0.62 – 1.10) |
| Vaginal mesh procedure | 0.95 (0.82 – 1.08) | 1.22 (1.10 – 1.35) | 1.04 (0.89 – 1.23) | 1.08 (0.91 – 1.28) | 1.34 (1.12 – 1.62) | 1.26 (1.03 – 1.55) |
| Concurrent sling procedure | 1.00 (0.89 – 1.12) | 1.15 (1.06 – 1.25) | 1.21 (1.06 – 1.38) | 1.19 (1.04 – 1.37) | 1.07 (0.91 – 1.26) | 1.04 (0.88 – 1.23) |
| Charlson Comorbidity Index Score | 1.13 (1.20 – 1.16) | 1.03 (1.01 – 1.06) | 1.14 (1.11 – 1.18) | STRATA | 0.99 (0.95 – 1.04) | 1.00 (0.96 – 1.05) |
The 90-day readmission rate was 7.3%. The most common reason for readmission was infection (2.0%), of which urinary tract infections (UTI) represented three quarters (Supplemental Table 1). Figure 1 displays the cumulative incidence for readmissions over 90 days using Kaplan-Meier estimates by Medicaid eligibility (p<.001) and concurrent sling procedure(p=0.004), respectively. After adjustment, women in the Northeast (aHR 0.7, 95%CI 0.6–0.9) were 30% less likely to be readmitted than women in the Midwest (Table 2). Medicaid eligibility (aHR 1.5, 95%CI 1.3–1.8) and concurrent sling procedures (aHR 1.2, 95%CI 1.04– 1.4) were associated with a higher risk of 90-day readmission.
Figure 1.
Readmission over 90 days
The 1-year reoperation rate was 4.5%. The most common type of reoperation was a sling procedure (1.8%) followed by sling revision surgery (0.9%). Figure 2 display the cumulative incidence for reoperations using Kaplan-Meier estimates by vaginal mesh (p=0.002) and type or POP surgery (p=0.001), respectively. After adjustment, vaginal mesh procedures (aHR 1.3, 95%CI 1.03 – 1.6) and laparoscopic sacrocolpopexy (aHR 1.4, 95%CI 1.1 – 1.8) were associated with higher risk of reoperation while obliterative POP surgery (aHR 0.6, 95%CI 0.4 – 0.9) was associated with a lower risk of reoperation (Table 2).
Figure 2.
Reoperation 1-year following index surgery
The median overall cost for minimally invasive POP surgery was $7,015 (interquartile range: $5,033 – $9,335) (Table 3). The overall cost for patients who experienced any adverse outcome was significantly higher than those who did not. Readmission was associated with the highest cost: ($16,390 vs no readmission: $6,777, p<.001). Overall cost differed significantly across type of POP repair with laparoscopic repair being most expensive (laparoscopic: $8,347 vs obliterative: $6,294 vs vaginal: $6,766, p<.001) (Table 4). Medicaid eligibility was associated with a higher cost than patients who were not Medicaid eligible ($7,441 vs $6,970, p<.001). Cost differed significantly across geographic regions (p<.001) with the South incurring the lowest cost ($6,637) and the Northeast incurring the highest costs ($7,838).
Table 3.
Cost associated with prolapse surgery across adverse outcomes
| Overall cohort N = 11,779 |
No complication N = 10,372 |
Complication N = 1,407 |
p-value | No Reoperation N = 11,196 |
Reoperation N = 583 |
p | No Readmission n = 10,886 |
Readmission n = 893 |
p-value | |
|---|---|---|---|---|---|---|---|---|---|---|
| Total cost | 7,015 [5,033 – 9,335] | 6,830 [4,912 – 8,940] | 8,962 [6,501 – 1,3251] | <.001 | 6,777 [4,873 – 8,754] | 16,390 [11,095 – 26,217] | <.001 | 6,968 [5,006 – 9,217] | 8,024 [6,013 – 11,372] | <.001 |
| Carrier | 1,934 [1,434 – 2,562] | 1,896 [1,408 – 2,493] | 2,289 [1,697 – 3,088] | <.001 | 1,880 [1,398 – 2,451] | 3,225 [2,276 – 4,597] | <.001 | 1,919 [1,426 – 2,535] | 2,247 [1,648 – 3,183] | <.001 |
| Inpatient | 0 [0 – 4,772] | 0 [0 – 0] | 4,878 [0 – 8,669] | <.001 | 0 [0 – 0] | 9,332 [5,700 – 16,186] | <.001 | 0 [0 – 4,741] | 0 [0 – 5,261] | 0.045 |
| Outpatient | 3,313 [336 – 5,002] | 3,453 [674 – 5,049] | 809 [30 – 4,245] | <.001 | 3,393 [464 – 5,027] | 1,149 [52 – 4,456 | <.001 | 3,298 [316 – 4,966] | 3,792 [1,071 – 5,783] | <.001 |
Table 4.
Cost across other variables of interest
| Laparoscopic N = 1706 |
Obliterative N = 1126 |
Vaginal N = 8713 |
p value | Not Medicaid Eligible | Medicaid Eligible | p value | Midwest | Northeast | South | West | p value | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total Cose | 8,347 [6,968 – 10,547] | 6,294 [4,277 – 9,898] | 6,766 [4,880 – 8,980] | <.001 | 6,970 [4,992 – 9,179] | 7,441 [5,328 – 10,656] | <.001 | 6,882 [4,968 – 8,952] | 7,838 [5,634 – 10,309] | 6,637 [4,767 – 8,658] | 7,801 [5,366 –10,458] | <.001 |
| Carrier | 2,282 [1,825 – 2,833] | 1,634 [1,168 – 2,316] | 1,905 [1,419 – 2,527] | <.001 | 1,924 [1,429 – 2,541] | 1,991 [1,493 – 2,659] | <.001 | 1,844 [1,365 – 2,413] | 2,038 [1,548 – 2,631] | 1,885 [1,402 – 2,514] | 2,115 [1,550 – 2,814] | <.001 |
| Inpatient | 0 [0 – 4,731] | 0 [0 – 5,547] | 0 [0 – 4,784] | 0.001 | 0 [0 – 4,659] | 0 [0 – 5,491] | <.001 | 0 [0 – 5,145] | 0 [0 – 6,198] | 0 [0 – 0] | 0 [0 – 5,748] | 0.024 |
| Outpatient | 5,076 [5011 – 6,254] | 2,791 [333 – 3,904] | 3,194 [265 – 4,710] | <.001 | 3,295 [291 – 4,988] | 3,416 [585 – 5,042] | 0.046 | 3,160 [322 – 4,826] | 3,399 [212 – 5,334] | 3,307 [795 – 4,877] | 3,424 [179 – 5,523] | <.001 |
Discussion:
In this study, we describe adverse outcomes following minimally invasive POP surgery in a US population of women 65 years of age and older. In this cohort of older patients, vaginal hysterectomy at the time of minimally invasive POP repair was associated with the highest likelihood of complications at the time of surgery. Medicaid eligibility was associated with the highest risk for readmission over 90-days, and obliterative surgery was associated with the lowest risk of reoperation over 1-year. Costs were significantly higher for patients who experienced an adverse outcome, with readmission associated with $9,613 increased cost compared to patients who were not readmitted. Laparoscopic SCP was on average $2,053 more costly than obliterative surgery and $1,581 more costly than vaginal POP repair.
In our study, vaginal hysterectomy was the factor most strongly associated with an increased likelihood of experiencing a complication at the time of the index procedure. Patients undergoing vaginal hysterectomy at the time of their POP repair had a 2.4 times higher likelihood of a complication compared to patients who did not undergo a hysterectomy. Only one-third of patients in our cohort underwent concurrent hysterectomy (20% vaginal and 12% laparoscopic). Dallas and colleagues in 2018 reported a similar proportion of women undergoing hysterectomy as part of their POP repair (45%) and found an association between hysterectomy and increased odds of both complications and readmission [26], though the authors did not distinguish between hysterectomy route. Erekson and colleagues in 2017 found no difference in 30-day major postoperative complications across routes of surgery using a different national database [27]; however, they did find concurrent hysterectomy, obliterative surgery and concurrent sling procedures to be associated with an increased risk of complications. This is consistent with our findings.
Regarding readmission, our rate of 90-day readmission was 7.3%. The most common reason for readmission was infection-related with the majority due to UTIs. Hokenstad and colleagues in 2017 reported a 2% readmission rate after POP surgery and described surgical site infections as the most common reason for readmission followed by non-surgical site infections[28]. Clancy and colleagues more recently in 2021 also reported a 2% rate of readmission after POP surgery. While they found surgical site infection as the most common reason for readmission following vaginal procedures, they identified gastrointestinal issues as the most common reason for readmission following laparoscopic procedures [29]. Our rate of readmission was higher than these two studies, at 7.3%; however, it is important to note that our readmission timeframe was 90-days rather than the 30-day timeframe used in the studies by Hokenstad and Clancy. Also, the rate of readmission for UTI was higher in our study compared to those by Hokenstad and Clancy. This difference may be due to a couple factors. First, the criteria for assigning the diagnosis of a UTI is standardized in the database used by both Hokenstad and Clancy, while the same diagnostic criteria are not required to associate a UTI diagnosis in the Medicare database. Second, as we utilized Medicare data, the mean age of our cohort was 72 years, which reflects an older cohort than those examined by Hokenstad (60 years) and Clancy (62 years), and older age has been identified as amongst the strongest predictors of postoperative UTI [30].
We found that POP procedures involving mesh, both laparoscopic SCP and vaginal mesh procedures, were associated with an increased risk of reoperation at 1 year. There are several published studies that describe an increased reoperation rate for mesh-augmented procedures[4],[10],[11]. Vandendriessche and colleagues reported the reoperation rates following SCP to be 12.5% over a mean of 4 years, and the most common indication for reoperation was urinary incontinence-related. Similarly, the most common reason for reoperation in our cohort was sling placement for treatment of stress urinary incontinence, followed by sling revision; however, just as in Vandendriessche study, we cannot comment whether this was due to urinary retention, persistent stress urinary incontinence or mesh exposure or extrusion. Dallas and colleagues reported a higher rate of repeat surgery in women who underwent mesh-augmented POP repair, but after controlling for confounding factors, they did not find that mesh was independently associated with reoperation. Similarly, after controlling for confounders, we established that mesh augmented POP repairs were only weakly associated with reoperation, and a very small proportion of cases were for mesh excision – vaginal or laparoscopic – though this may be indicative of the follow-up duration of 1-year. While in our study, hysterectomy was not independently associated with 1-year reoperation risk, several studies have commented on hysterectomy at the time of POP surgery as protective for reoperation[4],[5],[31].
Our findings add to the literature regarding sociodemographic differences in outcomes following minimally invasive POP surgery. We found that Medicaid eligibility was associated with a 50% increased risk of readmission. Several studies have described an association between low socioeconomic status and increased adverse events. Daugbjerg and colleagues described higher odds of complications in women of low socioeconomic status undergoing hysterectomy [32] as well as differences in surgical approach for hysterectomy based on socioeconomic status.[33] Dallas and colleagues reported Medicaid eligibility as an independent risk factor for unplanned hospital visit.[34]
We also report adverse outcomes following minimally invasive POP surgery differed across geographic regions and socioeconomic status even after controlling for confounding variables. Compared to the Midwest, the South was associated with a 25% decreased odds of complication, while the West was associated with a 13% increased odds of complication. The Northeast was associated a 28% decreased risk of readmission. Cost also differed across geographic regions with the Northeast incurring the highest cost. This finding differ from those published by Sheyn and colleagues who explored costs associated with benign hysterectomy [35] and reported highest cost in the West. These differences in adverse events and cost may be due to differences in practice patterns across the US. Medicare beneficiaries with low socioeconomic status are at higher risk of 90-day readmission following minimally invasive surgery for POP and such readmissions incur a nearly $10,000 increased cost. Given that the most common reason for readmission was UTI, targeted interventions to prevent UTIs in this older population of women in the postoperative period beyond 30-days could improve patient outcomes and reduce costs. To date, few studies have explored factors such as socioeconomic status and regional differences in practice patterns and outcomes through a healthcare equity and quality lens in the field of gynecology. As more studies exposes the link between healthcare disparities and adverse outcomes, more research is needed to explore ways to bridge gaps in care for marginalized individuals. We also need to investigate the drivers of differences seen in both cost and postoperative outcomes across the US as an important opportunities to mitigate cost and improve quality.
Strengths of this study include the use of the Medicare 5% LDS as it is the largest collection of longitudinal hospital care data in the US. We included all routes of minimally invasive surgery for POP. Limitations of this study include the fact given the nature of the Medicare program, we only were able to evaluate women 65 years and older and thus excluded a large portion of women who undergo surgery for POP in the US. Other limitations include those inherent to claims-based analyses such as the retrospective nature of data and potential coding misclassification. Claims-based analyses are also constrained by a lack of granular clinical details such as body mass index, stage of POP, past surgical history, and surgeon volume. We also did not distinguish between robotic-assisted laparoscopic surgery and traditional laparoscopic surgery. In terms of mesh related complications, we did not capture mesh excisions performed in the office or managed conservatively with vaginal estrogen.
In summary, we found that low socioeconomic status and route of surgery route of surgery are associated with adverse outcomes following minimally invasive surgery for POP in US women 65 years and older. UTI was the most common reason for readmission 90-days postoperatively, while surgery to treat stress incontinence and sling revisions were the most common reasons for a return to the operating room in the 1-year following the index surgery. Given that current predictions estimate that the US population aged 65 and older will be 90 million by the year 2050, and given that Medicare will be the primary insurance for the large majority of those older Americans, we need to continue to investigate drivers of adverse outcomes following elective POP surgery. While findings of this study can inform counseling of older women (≥65 years) regarding surgical risks, more research is needed to understand the socioeconomic drivers that place Medicaid eligible patients at a higher risk for readmission in the 90-days following surgery. Targeted interventions to prevent UTIs in this older population should be considered beyond 30-day postoperative period in order to improve patient outcomes and reduce costs.
Supplementary Material
Disclosures:
Authors do not have any relevant disclosures. C.E.B. is a consultant for Boston Scientific. D.S. is a consultant for Renalis and receives research funding from NICHD.
Financial support:
NICHD R25-HD094667 AUGS/Duke UrogynCREST (Urogynecology Clinical Research Educational Scientist Training) program
Contributor Information
C. Emi BRETSCHNEIDER, Division of Female Pelvic Medicine and Reconstructive Surgery, Departments of Obstetrics & Gynecology, Northwestern University Feinberg School of Medicine, Chicago, IL.
Charles D. SCALES, Jr, Departments of Surgery (Urology) and Population Health Science, Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC.
Oyomoare OSAZUWA-PETERS, Department of Population Health Science, Duke University School of Medicine, Durham, NC.
David SHEYN, Division of Female Pelvic Medicine and Reconstructive Surgery, Urology Institute, University Hospitals, Cleveland, OH.
Vivian SUNG, Department Obstetrics and Gynecology, Alpert Medical School at Brown University.
References:
- 1.Nygaard I, Barber MD, Burgio KL, Kenton K, Meikle S, Schaffer J, Spino C, Whitehead WE, Wu J, Brody DJ, Pelvic Floor Disorders N (2008) Prevalence of symptomatic pelvic floor disorders in US women. JAMA 300 (11):1311–1316. doi: 10.1001/jama.300.11.1311 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wu JM, Hundley AF, Fulton RG, Myers ER (2009) Forecasting the prevalence of pelvic floor disorders in U.S. Women: 2010 to 2050. Obstet Gynecol 114 (6):1278–1283. doi: 10.1097/AOG.0b013e3181c2ce96 [DOI] [PubMed] [Google Scholar]
- 3.Greer JA, Northington GM, Harvie HS, Segal S, Johnson JC, Arya LA (2013) Functional status and postoperative morbidity in older women with prolapse. J Urol 190 (3):948–952. doi: 10.1016/j.juro.2013.03.004 [DOI] [PubMed] [Google Scholar]
- 4.Vandendriessche D, Sussfeld J, Giraudet G, Lucot JP, Behal H, Cosson M (2017) Complications and reoperations after laparoscopic sacrocolpopexy with a mean follow-up of 4 years. Int Urogynecol J 28 (2):231–239. doi: 10.1007/s00192-016-3093-6 [DOI] [PubMed] [Google Scholar]
- 5.Mairesse S, Chazard E, Giraudet G, Cosson M, Bartolo S (2020) Complications and reoperation after pelvic organ prolapse, impact of hysterectomy, surgical approach and surgeon experience. Int Urogynecol J 31 (9):1755–1761. doi: 10.1007/s00192-019-04210-6 [DOI] [PubMed] [Google Scholar]
- 6.Bretschneider CE, Nieto ML, Geller EJ, Palmer MH, Wu JM (2016) The Association of the Braden Scale Score and Postoperative Morbidity Following Urogynecologic Surgery. Urol Nurs 36 (4):191–197 [PubMed] [Google Scholar]
- 7.Clancy AA, Gauthier I, Ramirez FD, Hickling D, Pascali D (2019) Predictors of sling revision after mid-urethral sling procedures: a case-control study. BJOG 126 (3):419–426. doi: 10.1111/1471-0528.15470 [DOI] [PubMed] [Google Scholar]
- 8.Weltz V, Guldberg R, Larsen MD, Lose G (2021) Body mass index influences the risk of reoperation after first-time surgery for pelvic organ prolapse. A Danish cohort study, 2010–2016. Int Urogynecol J 32 (4):801–808. doi: 10.1007/s00192-020-04482-3 [DOI] [PubMed] [Google Scholar]
- 9.Ringel NE, de Winter KL, Siddique M, Marczak T, Kisby C, Rutledge E, Soriano A, Samimi P, Schroeder M, Handler S, Zeymo A, Gutman RE (2021) Surgical Outcomes in Urogynecology-Assessment of Perioperative and Postoperative Complications Relative to Preoperative Hemoglobin A1c-A Fellow’s Pelvic Research Network Study. Female Pelvic Med Reconstr Surg. doi: 10.1097/SPV.0000000000001057 [DOI] [PubMed] [Google Scholar]
- 10.Dallas KB, Rogo-Gupta L, Elliott CS (2018) What Impacts the All Cause Risk of Reoperation after Pelvic Organ Prolapse Repair? A Comparison of Mesh and Native Tissue Approaches in 110,329 Women. J Urol 200 (2):389–396. doi: 10.1016/j.juro.2018.02.3093 [DOI] [PubMed] [Google Scholar]
- 11.Chughtai B, Barber MD, Mao J, Forde JC, Normand ST, Sedrakyan A (2017) Association Between the Amount of Vaginal Mesh Used With Mesh Erosions and Repeated Surgery After Repairing Pelvic Organ Prolapse and Stress Urinary Incontinence. JAMA Surg 152 (3):257–263. doi: 10.1001/jamasurg.2016.4200 [DOI] [PubMed] [Google Scholar]
- 12.Pacquée S, Nawapun K, Claerhout F, Werbrouck E, Veldman J, Dʼhoore A, Wyndaele J, Verguts J, De Ridder D, Deprest J (2019) Long-Term Assessment of a Prospective Cohort of Patients Undergoing Laparoscopic Sacrocolpopexy. Obstet Gynecol 134 (2):323–332. doi: 10.1097/aog.0000000000003380 [DOI] [PubMed] [Google Scholar]
- 13.Morling JR, McAllister DA, Agur W, Fischbacher CM, Glazener CM, Guerrero K, Hopkins L, Wood R (2017) Adverse events after first, single, mesh and non-mesh surgical procedures for stress urinary incontinence and pelvic organ prolapse in Scotland, 1997–2016: a population-based cohort study. Lancet 389 (10069):629–640. doi: 10.1016/S0140-6736(16)32572-7 [DOI] [PubMed] [Google Scholar]
- 14.Malacarne Pape D, Escobar CM, Agrawal S, Rosenblum N, Brucker B (2021) The impact of concomitant mid-urethral sling surgery on patients undergoing vaginal prolapse repair. Int Urogynecol J 32 (3):681–685. doi: 10.1007/s00192-020-04544-6 [DOI] [PubMed] [Google Scholar]
- 15.Diwadkar GB, Barber MD, Feiner B, Maher C, Jelovsek JE (2009) Complication and reoperation rates after apical vaginal prolapse surgical repair: a systematic review. Obstet Gynecol 113 (2 Pt 1):367–373. doi: 10.1097/AOG.0b013e318195888d [DOI] [PubMed] [Google Scholar]
- 16.Le Teuff I, Labaki M, Fabbro-Peray P, Debodinance P, Jacquetin B, Marty J, Letouzey V, Eglin G, de Tayrac R (2019) Perioperative morbi-mortality after pelvic organ prolapse surgery in a large French national database from gynecologist surgeons. J Gynecol Obstet Hum Reprod 48 (7):479–487. doi: 10.1016/j.jogoh.2019.05.008 [DOI] [PubMed] [Google Scholar]
- 17.Chughtai B, Mao J, Matheny ME, Mauer E, Banerjee S, Sedrakyan A (2021) Long-Term Safety with Sling Mesh Implants for Stress Incontinence. J Urol 205 (1):183–190. doi: 10.1097/ju.0000000000001312 [DOI] [PubMed] [Google Scholar]
- 18.Sung VW, Rogers ML, Myers DL, Clark MA (2006) Impact of hospital and surgeon volumes on outcomes following pelvic reconstructive surgery in the United States. Am J Obstet Gynecol 195 (6):1778–1783. doi: 10.1016/j.ajog.2006.07.015 [DOI] [PubMed] [Google Scholar]
- 19.Mehta A, Xu T, Hutfless S, Makary MA, Sinno AK, Tanner EJ 3rd, Stone RL, Wang K, Fader AN (2017) Patient, surgeon, and hospital disparities associated with benign hysterectomy approach and perioperative complications. Am J Obstet Gynecol 216 (5):497 e491–497 e410. doi: 10.1016/j.ajog.2016.12.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Projected Future Growth of Older Population. (2019). https://acl.gov/aging-and-disability-in-america/data-and-research/projected-future-growth-older-population. Accessed April 10, 2022
- 21.Dallas K, Dubinskaya A, Andebrhan SB, Anger J, Rogo-Gupta LJ, Elliott CS, Ackerman AL (2021) Racial Disparities in Outcomes of Women Undergoing Myomectomy. Obstet Gynecol 138 (6):845–851. doi: 10.1097/AOG.0000000000004581 [DOI] [PubMed] [Google Scholar]
- 22.Dällenbach P, Jungo Nancoz C, Eperon I, Dubuisson JB, Boulvain M (2012) Incidence and risk factors for reoperation of surgically treated pelvic organ prolapse. Int Urogynecol J 23 (1):35–41. doi: 10.1007/s00192-011-1483-3 [DOI] [PubMed] [Google Scholar]
- 23.Brusselaers N, Lagergren J (2017) The Charlson Comorbidity Index in Registry-based Research. Methods Inf Med 56 (5):401–406. doi: 10.3414/ME17-01-0051 [DOI] [PubMed] [Google Scholar]
- 24. [April 10, 2022]. https://www.medicaid.gov/medicaid/eligibility/index.html. Accessed.
- 25.Services CfMM (2018) Global Surgery Booklet.
- 26.Dallas K, Elliott CS, Syan R, Sohlberg E, Enemchukwu E, Rogo-Gupta L (2018) Association Between Concomitant Hysterectomy and Repeat Surgery for Pelvic Organ Prolapse Repair in a Cohort of Nearly 100,000 Women. Obstet Gynecol 132 (6):1328–1336. doi: 10.1097/aog.0000000000002913 [DOI] [PubMed] [Google Scholar]
- 27.Erekson E, Murchison RL, Gerjevic KA, Meljen VT, Strohbehn K (2017) Major postoperative complications following surgical procedures for pelvic organ prolapse: a secondary database analysis of the American College of Surgeons National Surgical Quality Improvement Program. Am J Obstet Gynecol 217 (5):608 e601–608 e617. doi: 10.1016/j.ajog.2017.05.052 [DOI] [PubMed] [Google Scholar]
- 28.Hokenstad ED, Glasgow AE, Habermann EB, Occhino JA (2017) Readmission and Reoperation After Surgery for Pelvic Organ Prolapse. Female Pelvic Med Reconstr Surg 23 (2):131–135. doi: 10.1097/SPV.0000000000000379 [DOI] [PubMed] [Google Scholar]
- 29.Clancy AA, Chen I, Pascali D, Minassian VA (2021) Surgical approach and unplanned readmission following pelvic organ prolapse surgery: a retrospective cohort study using data from the National Surgical Quality Improvement Program Database (NSQIP). Int Urogynecol J 32 (4):945–953. doi: 10.1007/s00192-020-04505-z [DOI] [PubMed] [Google Scholar]
- 30.Sanaee MS, Pan K, Lee T, Koenig NA, Geoffrion R (2020) Urinary tract infection after clean-contaminated pelvic surgery: a retrospective cohort study and prediction model. Int Urogynecol J 31 (9):1821–1828. doi: 10.1007/s00192-019-04119-0 [DOI] [PubMed] [Google Scholar]
- 31.Dallas KB, Trimble R, Rogo-Gupta L, Elliott CS (2018) Care Seeking Patterns for Women Requiring a Repeat Pelvic Organ Prolapse Surgery Due to Native Tissue Repair Failure Compared to a Mesh Complication. Urology 122:70–75. doi: 10.1016/j.urology.2018.08.017 [DOI] [PubMed] [Google Scholar]
- 32.Daugbjerg SB, Cesaroni G, Ottesen B, Diderichsen F, Osler M (2014) Effect of socioeconomic position on patient outcome after hysterectomy. Acta Obstet Gynecol Scand 93 (9):926–934. doi: 10.1111/aogs.12444 [DOI] [PubMed] [Google Scholar]
- 33.Daugbjerg SB, Ottesen B, Diderichsen F, Frederiksen BL, Osler M (2012) Socioeconomic factors may influence the surgical technique for benign hysterectomy. Dan Med J 59 (6):A4440. [PubMed] [Google Scholar]
- 34.Dallas KB, Sohlberg EM, Elliott CS, Rogo-Gupta L, Enemchukwu E (2017) Racial and Socioeconomic Disparities in Short-term Urethral Sling Surgical Outcomes. Urology 110:70–75. doi: 10.1016/j.urology.2017.08.021 [DOI] [PubMed] [Google Scholar]
- 35.Sheyn D, Mahajan S, Billow M, Fleary A, Hayashi E, El-Nashar SA (2017) Geographic Variance of Cost Associated With Hysterectomy. Obstet Gynecol 129 (5):844–853. doi: 10.1097/AOG.0000000000001966 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.