Objective
The objective of the study was to examine the trends in the rate of peripartum hysterectomy and the contribution of changes in maternal characteristics to these trends.
Study Design
This was a cross-sectional study of peripartum hysterectomy identified from hospitalizations for delivery recorded in the 1994-2007 Nationwide Inpatient Sample.
Results
The overall rate of peripartum hysterectomy increased by 15% during the study period. The rate of hysterectomy for abnormal placentation increased by 1.2-fold; adjustment for previous cesarean delivery explained nearly all of this increase. The rate of hysterectomy for uterine atony following repeat cesarean delivery increased nearly 4-fold, following primary cesarean delivery approximately 2.5-fold, and following vaginal delivery about 1.5-fold. This fast growing trend in peripartum hysterectomy secondary to uterine atony was also largely explained by increasing rates of primary and repeat cesareans.
Conclusion
Rates of peripartum hysterectomy increased substantially in the United States from 1994 to 2007; much of this increase was due to rising rates of cesarean delivery.
Postpartum hemorrhage is one of leading causes of maternal morbidity and mortality worldwide. When severe obstetric hemorrhage has failed to respond to other treatment, peripartum hysterectomy is usually performed. Hence, peripartum hysterectomy can be considered an indicator of severe postpartum hemorrhage. Because it is associated with substantial maternal morbidity and results in the loss of fertility, understanding trends in this procedure and factors associated with it has the potential to improve patient safety and quality of care. The leading indications for peripartum hysterectomy include invasive placentation and uterine atony.
During the past decade, changing characteristics of the delivering population and in obstetric practice in the United States have increased the prevalence of known risk factors for conditions that lead to peripartum hysterectomy. The rate of cesarean delivery has risen dramatically ; the most important risk factor for invasive placentation is the number of prior cesarean deliveries. Risk factors for uterine atony have also become more prevalent among pregnant women including advanced maternal age, multiple gestations, hypertensive diseases of pregnancy, labor inductions, and cesarean delivery.
Recent US nationwide studies have shown increases in postpartum hemorrhage among delivery hospitalizations. Much of the information about trends and risk factors for peripartum hysterectomy is known from other developed countries and state-based and/or institution-based studies ; there is a paucity of US population-based information about this procedure. Understanding changes in the incidence of peripartum hysterectomy is important in assessing the consequences of changes in contemporary obstetric practice, particularly the rising rate of cesarean delivery. This study uses the largest database of US hospital admissions, the Nationwide Inpatient Sample (NIS), to examine trends in the incidence of peripartum hysterectomy from 1994 to 2007. The study also attempts to understand the extent to which trends in peripartum hysterectomy can be explained by changes in the characteristics of delivering patients.
Materials and Methods
The NIS is the largest nationwide all-payer hospital inpatient care database in the United States. It is one of a family of databases and software tools developed as part of the Healthcare Cost and Utilization Project sponsored by the Agency for Healthcare Research and Quality (AHRQ) in partnership with state-level data-collection organizations to provide national estimates for inpatient care.
The NIS is a stratified sample of approximately 20% of all US community hospitals, with hospitals selected using 5 characteristics (rural/urban location, number of beds, region, teaching status, and ownership) to create a sample that is maximally representative of all US hospitalizations. The database includes all discharges from selected hospitals and provides information on 5 million to 8 million discharges from approximately 1000 hospitals each year. Because the data are publicly available and do not contain direct personal identifiers, this study was exempt from review by institutional review boards.
Our analysis included all delivery hospitalizations identified in the 1994-2007 NIS. Delivery hospitalizations were identified by an enhanced method that incorporates delivery-related International Classification of Diseases , ninth revision, Clinical Modification (ICD-9-CM) and diagnosis-related groups codes. This method has been shown to improve the accuracy of identifying deliveries, especially those with severe complications. The ICD-9-CM codes were used to identify hospitalizations with peripartum hysterectomy, with peripartum hysterectomy being defined as hysterectomy occurring during the hospitalization for delivery.
Because peripartum hysterectomy is performed for many different reasons, including abnormal placentation that either causes or is expected to cause hemorrhage, uterine atony, and uterine rupture, we stratified our analysis of trends by the indication for hysterectomy. Because the risk factors for atony may vary by mode of delivery, we further stratified our analysis of hysterectomy with uterine atony by mode of delivery.
Therefore, the indication for the hysterectomy, if available, was assigned to 1 of 8 conditions, in hierarchical order: 1 of 3 conditions of hysterectomy with placental abnormalities (retained placenta with hemorrhage, retained placenta without hemorrhage, and placenta previa), hysterectomy with atony with repeat cesarean delivery, hysterectomy with atony with primary cesarean delivery, hysterectomy with atony with vaginal delivery, hysterectomy with delayed postpartum hemorrhage, and hysterectomy with uterine rupture.
The following risk factors for postpartum hemorrhage were also identified using ICD-9-CM codes: medical induction, mode of delivery (repeat cesarean, primary cesarean, vaginal after cesarean, and vaginal with no history of cesarean), fetal macrosomia, multiple births, preterm labor, chorioamnionitis, placental abruption, fibroids, eclampsia/severe preeclampsia, polyhydramnios, stillbirth, antenatal hemorrhage because of coagulation disorders, and prolonged labor.
Statistical analyses
All statistical analyses were conducted using SAS 9.2-callable SUDAAN (Research Triangle Institute, Research Triangle Park, NC) to account for the complex sampling design of the NIS. Rates per 100,000 deliveries were calculated for each of the 8 hysterectomy indications.
We also reported a prevalence (in percent) of medical induction, previous cesarean delivery, fetal macrosomia, multiple births, preterm labor chorioamnionitis, and placental abruption. Rates for infrequently coded risk factors for postpartum hemorrhage (fibroid, eclampsia/severe preeclampsia, polyhydramnios, stillbirths, antenatal hemorrhage because of coagulation disorders, prolonged labor) and in-hospital mortality were reported per 100,000 deliveries.
Orthogonal polynomial coefficients were obtained recursively by the method of Fisher and Yates to test linear trend across all 7 intervals: 1994-1995, 1996-1997, 1998-1999, 2000-2001, 2002-2003, 2004-2005, and 2006-2007. The differences in rates between 1994-1995 and 2006-2007 were tested by applying linear contrasts.
Logistic regression models were run to obtain odds ratios and 95% confidence intervals showing the likelihood of the 2 most prevalent hysterectomy indications: hysterectomy with placental abnormalities and hysterectomy with atony, with the latter stratified by mode of delivery (atony with primary cesarean delivery, repeat cesarean delivery, or vaginal delivery), using all delivery hospitalizations as the denominator.
In an effort to separate out the contribution of the changing prevalence of primary and repeat cesarean delivery to the trend, we also modeled trends in hysterectomy for atony considering each mode of delivery as a separate group (ie, we examined trends just within a given mode of delivery), using only those hospitalizations with comparable mode of delivery in the denominator. The reported odds ratios for 2006-2007 compared with 1994-1995 were adjusted for age, hospital region (Northwest, Midwest, South, West), insurance status, disposition, hospital location/teaching status (urban-teaching, urban nonteaching, and rural), and risk factors for peripartum hysterectomy described above.
Adjusted population-attributable fraction of deliveries with previous cesarean for peripartum hysterectomy with placental abnormalities for the 1994-1995 and 2006-2007 study periods were calculated, incorporating adjusted odds ratios and confidence limits using methods as described elsewhere.
Results
During the study period, there were an estimated 56,216,424 delivery discharge records. The overall rate of peripartum hysterectomy increased from 71.6 to 82.6 per 100,000 deliveries from 1994-1995 to 2006-2007 (or by 15%, P for linear trend < .001) ( Figure ) . The rate of hysterectomy with indication of abnormal placentation increased from 32.9 to 40.5 per 100,000 deliveries during this period (or by 23%, P for linear trend < .001). In contrast, hysterectomy for uterine atony increased from 11.2 to 25.9 per 100,000 deliveries (or by 130%, P for linear trend < .001). Hysterectomy performed for indications other than abnormal placentation, atony, delayed procedure for peripartum hysterectomy (PPH), and uterine rupture decreased during the period of study, from 20.4 to 13.6 per 100,000 deliveries. Detailed classification of peripartum hysterectomy and changes in rate of different type of hysterectomy are shown in Table 1 .
| Variable | ICD-9-CM code | Estimated n | 1994-1995 | 1996-1997 | 1998-1999 | 2000-2001 | 2002-2003 | 2004-2005 | 2006-2007 | P for linear trend |
|---|---|---|---|---|---|---|---|---|---|---|
| Total deliveries, n | 7,569,995 | 7,568,973 | 7,605,650 | 8,073,731 | 8,192,946 | 8,418,655 | 8,786,475 | |||
| Hysterectomy by indication | 124 a | Rate per 100,000 deliveries | ||||||||
| Retained placenta with hemorrhage | 666.0x | 12,557 | 19.6 (1.3) | 22.3 (1.6) | 19.3 (1.2) | 19.8 (1.2) | 22.5 (1.4) | 25.9 (1.4) | 26.1 (1.4) | ≤ .0001 |
| Retained placenta without hemorrhage | 667.0x, 667.1x | 2277 | 2.8 (0.5) | 3.3 (0.5) | 3.6 (0.5) | 3.7 (0.5) | 4.1 (0.6) | 4.5 (0.6) | 6.0 (0.6) | ≤ .0001 |
| Placenta previa | 641.0x, 641.1x | 5145 | 10.5 (1.2) | 8.6 (0.8) | 8.9 (0.8) | 9.6 (0.9) | 9.0 (0.7) | 9.1 (0.8) | 8.4 (0.7) | .30 |
| Atony, repeat CS delivery | 666.1x with 134 a + 654.2x | 2917 | 1.9 (0.4) | 3.8 (0.5) | 5.5 (0.7) | 4.8 (0.6) | 5.8 (0.6) | 6.5 (0.7) | 7.5 (0.7) | ≤ .0001 |
| Atony, primary CS delivery | 666.1x with 134 a | 4530 | 4.4 (0.6) | 4.8 (0.6) | 8.0 (0.8) | 9.0 (0.8) | 8.7 (0.8) | 9.7 (0.8) | 11.2 (0.9) | ≤ .0001 |
| Atony, vaginal delivery | 666.1x without 134 a | 4044 | 4.9 (0.6) | 5.6 (0.6) | 8.1 (0.7) | 8.7 (0.8) | 8.0 (0.7) | 7.8 (0.7) | 7.1 (0.6) | .003 |
| Delayed PPH | 666.2x | 648 | 1.8 (0.3) | 1.7 (0.3) | N/A | 1.2 (0.3) | 1.3 (0.3) | 0.9 (0.2) | 0.9 (0.2) | .01 |
| Uterine rupture | 665.0x, 665.1x | 1981 | 5.3 (0.6) | 5.1 (0.6) | 4.2 (0.6) | 3.5 (0.5) | 2.8 (0.4) | 2.5 (0.4) | 1.8 (0.3) | ≤ .0001 |
| Other/unspecified | N/A | 8918 | 20.4 (1.7) | 18.1 (1.6) | 15.7 (1.3) | 13.8 (1.0) | 15.3 (1.3) | 14.7 (1.1) | 13.6 (1.0) | ≤ .0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
