Database and requested parameters

This large retrospective cohort study was conducted with the use of data from the Canadian Institute of Health Information Discharge Abstract Database that captures demographic, administrative, and clinical information for all hospital discharges (inpatient and day surgeries) in Canada. Previous studies that validated the Discharge Abstract Database have reported a high degree of accuracy in the procedure codes and primary diagnosis codes. Ethics approval was obtained from the University of British Columbia Clinical Research Ethics Board. All women who underwent any or any combination of salpingectomy, hysterectomy, oophorectomy, fimbriectomy, or tubal ligation in the Canadian province of BC from Jan. 1, 2008 (before campaign initiation), to Dec. 31, 2011 (after the campaign and most recent complete calendar year data that were available at the time of our request), were included in this study. Patients who were <15 years old and patients who were not coded as being of female sex were excluded. Canadian Classification for Health Intervention codes were used to identify patients who underwent the surgical procedures of interest. A diagnosis code, International Classification of Diseases, 10th edition ( ICD-10)-CA Z30.2, indicated that the encounter was for sterilization purposes specifically. The Discharge Abstract Database provided information on operating time (time from first skin incision until completed skin closure), surgical approach (vaginal, laparoscopic, combined vaginal and laparoscopic, open), surgical indication, and length of hospital stay (LOS). Data were also gathered for patients who required blood transfusion and/or readmission to hospital, which reflected possible surgical complications. Parameters that were chosen for this study are of interest from an educational/knowledge translation perspective and were selected based on a provincial and cross-Canada survey of practicing gynecologists who identified concerns that were associated with the recommended change in surgical practice. This study will inform which patient groups are appropriate candidates for the preventative surgery and which surgical practices are feasible across health authorities of varying resources. All statistical analyses were performed with Stata software (version 12; StataCorp, College Station, TX).

Procedural uptake

The rates of salpingectomy between 2008-2011, which include the number of hysterectomies that were performed with and without BS or salpingo-oophorectomy and the number of sterilizations that were performed with BS or tubal ligation, were examined as measures of the baseline rates for the surgical procedures before the 2010 educational campaign and the uptake of the recommended change in practice after the campaign. χ ² analysis was performed to indicate whether there were significant differences in the rate of procedures across the 4-year time period.

Operative/perioperative measures

To investigate whether BS was associated with a higher risk of complications, data for women who underwent hysterectomy were divided into 3 categories that were based on surgical procedures received: (1) hysterectomy alone (the reference group, because these women were expected to be at lowest risk for complications), (2) hysterectomy with BS but no oophorectomy, and (3) hysterectomy with bilateral salpingo-oophorectomy (BSO). Comparisons were also made between the cohorts of women who had a diagnosis code that indicated that the encounter was for sterilization purposes and who underwent either (1) tubal ligation (the reference group, because this was standard practice for sterilization) or (2) isolated BS (defined as BS without accompanying hysterectomy or oophorectomy). Because the continuous variables were distributed normally, data were presented as means with standard deviations. Differences in the age of patients, OR time, LOS, hospital readmission, and the rate of blood transfusion were analyzed by χ ² tests for categoric variables and independent samples t tests for continuous variables. We ran logistic regressions, controlling for patient age as a potential confounder and cesarean section within that hospital stay among women who underwent tubal ligation and salpingectomy (because these procedures frequently occurred in the same hospital stay and both influence the rate of complications). We obtained aORs for the risk of hospital readmission and blood transfusion. All of these parameters were also compared for each year of the study period in cohorts of women who underwent hysterectomy with BS or isolated BS that was performed by different surgical approaches that included open procedure (the reference category), laparoscopic, vaginally, or using a combined (laparoscopic and vaginal) approach and among women who underwent isolated BS by open procedure (the reference category), laparoscopic, or vaginal approach.

Regional variation

There are 16 Health Service Delivery Areas (HSDAs) in BC that are based on geography and population distribution. We were interested to see whether the educational campaign had influenced all regions equally or whether there might be isolated areas where knowledge translation had failed. We investigated the regional variation in the rates of hysterectomy alone or in combination with BS or BSO and the rates of isolated salpingectomy across HSDAs as well as regional differences in surgical approach for these procedures.

Procedural uptake

There were 43,973 women who underwent ≥1 of our requested surgical procedures. Of these women, 4 were dropped because they were not coded as female, and 38 were dropped because they were <15 years old, which left us with a total study population of 43,931 women. In our sample, there were no missing data on primary procedure, primary diagnostic code, or age. Women for whom data on OR time were missing were excluded from the descriptive analysis of mean OR time. In this study population 21,003 women underwent hysterectomy; 21,411 women underwent BS; 13,719 women underwent tubal ligation, and 15,285 women underwent oophorectomy. Although procedures were not mutually exclusive, women were only included in 1 group based on the procedures that they had received (eg, a woman who underwent a hysterectomy with salpingectomy is included only in the group of women who underwent hysterectomy with salpingectomy and not also included in the hysterectomy group and salpingectomy group). Women in our sample had a mean age of 44.7 years.

The share of hysterectomies that were performed with salpingectomy increased significantly between 2008 and 2011 ( P < .001; Figure 1 ). Although only 45% of hysterectomies included removal of the fallopian tubes (BS or BSO) in 2008, this number had increased to 79% in 2011, which represented a statistically significant increase in the number of hysterectomies with salpingectomies across our study period ( P < .001).

Specific procedures that were performed from 2008-2011 in British Columbia

Proportion of women who underwent hysterectomy alone ( green ), hysterectomy with bilateral salpingectomy ( red ), and hysterectomy with bilateral salpingo-oophorectomy ( blue ).

McAlpine. Uptake and risks of opportunistic salpingectomy. Am J Obstet Gynecol 2014 .

The rate of hysterectomy with BS (without oophorectomy) increased from 5% of all hysterectomy procedures in 2008 to 35% in 2011 ( P < .001). The greatest change was observed after the September 2010 campaign; 80.7% of all hysterectomies with salpingectomy that were performed in 2010 were performed between Sept. 1 and Dec. 31, 2010. The numbers of radical hysterectomies were relatively stable over the time period (mean, 73) but radical hysterectomies with BS (and ovarian preservation) increased from 9% before the campaign to 24% after the campaign.

Figure 2 shows the share of patients who underwent tubal ligation or isolated salpingectomy who had a diagnosis code that indicated that the encounter was for sterilization. Although very few isolated BSs (0.5%) were being performed for those with a diagnosis of sterilization in 2008 and 2009, by 2011, 33.3% of sterilizations were done with the use of isolated BS, which represented a statistically significant increase in the use of salpingectomy for sterilization ( P < .001). Again a remarkable difference was observed after the September 2010 campaign because 98.1% of all salpingectomies for sterilization that were done in 2010 were done after the September campaign.

Procedures with a diagnosis code that indicated the encounter was for sterilization that were performed from 2008-2011 in British Columbia

Proportion of women who underwent isolated bilateral salpingectomy ( red ) or tubal ligation ( blue ).

McAlpine. Uptake and risks of opportunistic salpingectomy. Am J Obstet Gynecol 2014 .

The number of hysterectomies with BS that were performed with a diagnosis code that indicated prophylactic (risk-reducing) surgery (ICD-10-CA Z40.0, Z40.8, or Z40.9) in years 2008-2011 were 1, 1, 106, and 152, respectively (a statistically significant increase over time; P < .001) and isolated BS carried out with a prophylactic code also increased 1, 0, 52, and 97, respectively, across the 4 study years ( P < .001). In contrast, the corresponding numbers of prophylactic BSOs over the study period did not change at 64, 54, 64, and 72 over the 4 years ( P = .320). The breakdown of these procedures by patient age is illustrated in Table 1 and Figure 3 . The number of women who had a hysterectomy with BS increased in all age groups, particularly in women who were <50 years old ( Figure 3 , B; P < .001); this age cohort also had an increased number of isolated salpingectomies in 2011 compared with 2008 ( P < .001; Figure 3 , D). The number of women who underwent fimbriectomy only (without hysterectomy, oophorectomy, or salpingectomy) also increased over the study period ( P = .019); this procedure was most commonly performed in women who were <40 years old ( Figure 3 , E).

Number of women in British Columbia who underwent a procedure from 2008-2011 distributed according to patient age at the time of the surgery

A , Hysterectomy alone (n = 8362 women); B , hysterectomy with bilateral salpingectomy (n = 3670 women); C , hysterectomy with bilateral salpingo-oophorectomy (n = 8904 women); D , isolated salpingectomy (n = 2504 women); E , fimbriectomy (n = 1068 women). 15-39 years old, orange ; 40-44 years old, blue ; 45-49 years old, green ; 50-54 years old, purple ; ≥55 years old, red .

McAlpine. Uptake and risks of opportunistic salpingectomy. Am J Obstet Gynecol 2014 .

Operative/perioperative measures

Mean OR time was significantly longer in the hysterectomy with BS group ( P < .001) and the hysterectomy with BSO group ( P < .001) and compared with hysterectomy alone ( Table 2 ); however, the differences were only 16.3 minutes and 22.4 minutes on average for each group, respectively. Table 2 also shows that LOS among those women who were discharged home after hysterectomy alone (mean LOS, 2.52 days) was slightly longer than among those who underwent hysterectomy with BS (mean LOS, 2.37 days; P = .010). There were no significant differences in rates of blood transfusion across the 3 groups of hysterectomy patients; approximately 2.5% of patients received a blood transfusion in all groups. There were significant differences across the groups in rates of hospital readmission; patients who had a hysterectomy with BSO had a higher rate of readmission to the hospital (5.7% compared with 4.5% for hysterectomy alone; P < .001), but this was not observed for hysterectomy with BS ( P = .632, no difference from hysterectomy alone).

Mean OR time for sterilization by BS was longer by 10.2 minutes than tubal ligation (mean, 61.0 vs 71.2 minutes; P < .001). There were no statistically significant differences in LOS among patients who underwent salpingectomy for sterilization (mean LOS, 1.23 days) as compared with tubal ligation (mean LOS, 1.23 days; P = .117). Although both procedures are typically outpatient procedures, both groups contained women who stayed for postpartum care after live birth (41.1% of women in the tubal ligation group and 37.2% of women in the salpingectomy for sterilization group). The mean LOS decreases to 0.11 and 0.10 days when women in the postpartum period are removed from the tubal ligation and salpingectomy for sterilization groups, respectively ( Table 2 ).

We wanted to examine differences in crude odds ratios to determine whether there were crude differences in complication rates among patients who undergo salpingectomy to ascertain any high-level difference in safety profiles among the procedures. However, the age group of women differed significantly (as expected) and women who underwent tubal ligation and isolated salpingectomy were much more likely to have just undergone delivery by cesarean section. Thus, we calculated age-adjusted odds ratios (aORs) for the hysterectomy groups and controlled for cesarean delivery in the tubal ligation and isolated salpingectomy groups to examine the risk of readmission to the hospital and blood transfusion ( Table 2 ). Women who had a hysterectomy with BS were at no increased risk for readmission to the hospital or blood transfusion compared with those who had a hysterectomy alone (aOR, 0.91; 95% confidence interval [CI], 0.75–1.10 and aOR, 0.86; 95% CI, 0.67–1.10 for readmission and blood transfusion, respectively). With respect to women whose encounter was for sterilization purposes, women who underwent a BS were not at increased risk for either readmission or blood transfusion (aOR, 0.83; 95% CI, 0.56–1.23 and aOR, 0.77; 95% CI, 0.56–1.23, respectively).

Figure 4 shows the total number of hysterectomies with BS and the total number of isolated BSs that were performed by different surgical approaches across the study years. The proportion of hysterectomies with BS that were performed by open procedure decreased significantly from 77% in 2008 to 44% in 2011 ( P < .001; Figure 4 , A). For isolated salpingectomies, although the total number of these procedures increased dramatically from 2008-2011, there was no significant difference in surgical approach across time ( P = .127; Figure 4 , B).

Hysterectomies with BS and isolated BS by surgical approach from 2008-2011 in British Columbia

A , Hysterectomy with bilateral salpingectomy (n = 3670 women) and B , isolated bilateral salpingectomy (n = 2504 women) according to surgical approach (open procedure, blue ; laparoscopy, red ; vaginal, green ; combined approach, orange ). For hysterectomies with bilateral salpingectomy, the number of procedures that were performed by open laparotomy decreased significantly over the 4-year study period; there was a corresponding increase in laparoscopic, vaginal, and combined procedures ( P < .001 for all). For isolated bilateral salpingectomies, although the total number of procedures that were performed increased from 2008-2011, there was no significant difference in surgical approach across time ( P = .127).

McAlpine. Uptake and risks of opportunistic salpingectomy. Am J Obstet Gynecol 2014 .

Table 3 shows the differences in patient age, OR time, LOS and readmission, blood transfusion rates, and aORs for hysterectomies with BS, and isolated salpingectomies by surgical approach. OR time was significantly longer for laparoscopic approach and the combined approach than for open procedure (mean OR time, 168.1 and 155.3 minutes vs 124.7 minutes, respectively; all P < .001) and significantly shorter for the vaginal approach (mean OR time, 112.3 minutes; P < .001). Rates of readmission for hysterectomy with BS were lowest in those women who had a vaginal approach (2.4%; aOR, 0.51; 95% CI, 0.37–0.70), and combined modality also conferred a reduced risk of readmission (aOR, 0.71; 95% CI, 0.59–0.86).

Table 3

Operative/perioperative measures of risk of opportunistic salpingectomy by surgical approach

Variable	Open procedure	Laparoscopic	P value a	Vaginal	P value a	Combined vaginal and laparoscopic	P value a
Hysterectomy with bilateral salpingectomy, n	1885	426		677		682
Age, y b	43.1 ± 6.4	42.4 ± 6.1	.042	46.5 ± 10.6	< .001	42.4 ± 7.1	.015
Operating room time, min b	124.7 ± 42.7	168.1 ± 59.9	< .001	112.3 ± 43.7	< .001	155.3 ± 47.8	< .001
Missing data on operating room time, n	211	12	—	31	—	26	—
Length of hospital stay, d b	2.94 ± 1.7	1.58 ± 2.8	< .001	2.03 ± 1.9	< .001	1.60 ± 0.8	< .001
Readmission, n (%)	89 (4.7)	21 (4.9)	.847	16 (2.4)	.009	35 (5.1)	.739
Readmission, adjusted odds ratio c	1.00 (Reference)	0.86 (0.65, 1.13)	.232	0.51 (0.37, 0.70)	< .001	0.71 (0.59, 0.86)	.001
Blood transfusion, n (%) d	67 (3.6)	4 (0.9)	.005	3 (1.3)	.004	10 (1.5)	.007
Isolated salpingectomy	845	1647		12	0
Age, y b	37.2 ± 8.7	38.8 ± 7.3	< .001	42.8 ± 11.3	.027	—	—
Operating room time, min b	98.6 ± 52.9	84.5 ± 41.7	< .001	106.2 ± 70.0	.656	—	—
Missing data on operating room time	233	130	—	2	—	—	—
Length of hospital stay, d b	3.34 ± 6.5	0.33 ± 0.97	< .001	1.75 ± 2.45	.375	—	—
Readmission, n (%)	38 (4.5)	36 (2.2)	.001	0	.452	—	—
Readmission, adjusted odds ratio c	1.00 (Reference)	0.62 (0.54, 0.71)	< .001	0.50 (0.38, 0.65)	< .001	—	—
Blood transfusion, n (%) d	18 (2.1)	3 (0.2)	< .001	0	.609	—	—

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. The contribution of genetic and environmental factors to the duration of pregnancy
2. Brainstem and cerebellar differences and their association with neurobehavior in term small-for-gestational-age fetuses assessed by fetal MRI
3. Impact of monozygotic twinning on multiple births resulting from in vitro fertilization in the United States, 2006–2010
4. The forgotten hysterectomy

Stay updated, free articles. Join our Telegram channel

Join