Clinical analysis of emergency exploratory laparotomy in patients with intractable postpartum hemorrhage.

OBJECTIVE: This study was performed to explore the causes and clinical characteristics of emergency exploratory laparotomy in patients with intractable postpartum hemorrhage.
METHODS: This retrospective study was performed from January 2004 to December 2017. Patients with intractable postpartum hemorrhage were grouped according to the initial pathogenesis as determined by exploratory laparotomy: uterine atony, placental factors, coagulation dysfunction, or uterine rupture.
RESULTS: This study involved 72 patients who underwent emergency exploratory laparotomy, accounting for 0.04% of total deliveries. Uterine preservation surgery and hysterectomy were performed in 31 and 41 patients, respectively. Abnormal events upon returning to the ward were primarily vaginal hemorrhage and pelvic hematoma. The frequency of uterine artery ligation was lower in the hysterectomy group than uterine preservation group. The prothrombin activity level, fibrinogen level, and platelet count before surgery were lower in the hysterectomy group than uterine preservation group. The international normalized ratio and activated partial thromboplastin time were higher in the hysterectomy group than uterine preservation group. In total, 44 patients developed complications.
CONCLUSION: Placental implantation is a primary cause of hysterectomy after emergency laparotomy. However, the possibility of postpartum hemorrhage caused by coagulation disorders should not be ignored.

Introduction

Hemorrhage following the end of pregnancy is a severe obstetric complication. When conservative treatment for such postpartum hemorrhage is ineffective, emergency exploratory laparotomy is generally needed to save the life of the mother.[1] Although such cases rarely occur, they are associated with a poor prognosis; the surgery itself also has a high mortality rate, making it a serious medical emergency warranting the utmost medical attention.[2,3] Previous studies have shown that the incidence of exploratory relaparotomy after cesarean section ranges from 0.2% to 0.7%.[4,5]

Postpartum relaparotomy is an important risk factor for perinatal hysterectomy.[6,7] Hysterectomy can be used as a final approach to control postpartum hemorrhage and save the life of the mother when conservative treatments fail. Although hysterectomy can save the life of the patient in many instances, it is associated with long-term physiological and psychological impacts.[8] However, efforts to save the uterus may delay effective treatment, resulting in maternal death. Weighing the relative advantages and disadvantages of these two options while attempting to act in the best interests of the patient remains difficult for surgeons who treat intractable postpartum hemorrhage.

In this study, we retrospectively analyzed clinical data from patients who underwent emergency exploratory laparotomy for treatment of intractable postpartum hemorrhage at Fujian Provincial Maternity and Children’s Hospital from 2004 to 2017. We analyzed the relationship between clinical variables and the instances/outcomes of uterine preservation surgery and hysterectomy for these patients. The aim of this analysis was to explore the pathogenesis of and high-risk factors for uterine removal in these individuals.

Materials and methods

Study objective

This study included women who had completed delivery and developed intractable postpartum hemorrhage for which they underwent emergency exploratory laparotomy from January 2004 to December 2017 at Fujian Provincial Maternity and Children’s Hospital. The exclusion criteria were a gestation period of <28 weeks and performance of the exploratory laparotomy >6 weeks after delivery. This study was approved by the Institutional Ethics Committee of Fujian Provincial Maternity and Children’s Hospital, and all patients provided written informed consent.

Methods

This retrospective study was conducted in a Grade A referral center in southeastern China. Clinical data of all included patients were collected. The patients were divided into in a uterine preservation group and a hysterectomy group based on their surgical condition, pathogenesis, intraoperative exploratory findings, efforts made to prevent uterine removal, coagulation indexes, and complications. Blood loss was determined based on a volume method and a weighing method.

Statistical analysis

IBM SPSS Statistics for Windows, Version 23.0 (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Continuous variables with a normal distribution are presented as mean ± standard deviation and were compared by independent-sample t-tests. Continuous variables with a non-normal distribution are presented as median (range) and were compared by rank sum tests. Qualitative data are expressed as frequency and percentage and were compared by the chi-squared test or Fisher’s exact test. A P value of <0.05 was the threshold of statistical significance.

Results

Comparison of patients’ clinical characteristics

About 7,000 deliveries occurred in our hospital in 2004, and this number increased to 15,000 deliveries in 2017. A total of 72 patients underwent emergency exploratory laparotomy in our hospital from January 2004 to December 2017, accounting for 0.04% (72/168,860) of all deliveries during this same period. Among these 72 patients, successful conservative treatment was performed in 31 (43.1%) and hysterectomies were performed in 41 (56.9%); 2 patients who underwent a hysterectomy had been referred from other hospitals. The clinical characteristics of the patients in these two groups are compared in Table 1. There was no statistically significant difference in age, number of pregnancies, number of births, number of miscarriages, history of cesarean section, fetal position, or delivery mode between these two groups. With respect to obstetric complications, the proportion of placental implantation was significantly higher in the hysterectomy group than uterine preservation group (P = 0.015).

Table 1.

Patient characteristics.

	Hysterectomy(n = 41)	Uterine preservation (n = 31)	P
Age, years	31.66 ± 5.33	32.42 ± 5.49	0.636
Gravity	3.24 ± 1.52	2.74 ± 1.23	0.14
Parity	1.07 ± 0.61	0.87 ± 0.48	0.26
Prior abortion	1.14 ± 0.84	0.83 ± 0.53	0.26
Previous CS	16 (39.0)	12 (38.7)	0.978
Gestation weeks	35.71 ± 6.64	35.58 ± 6.78	0.93
Fetal birth weight	2969 ± 887	2867 ± 847	0.396
Fetal position
Vertex	31 (75.6)	26 (83.9)	0.393
Breech	9 (22.0)	2 (6.5)	0.139
Transverse	1 (2.4)	3 (9.7)	0.308
Mode of delivery
Vaginal delivery	9 (22.0)	6 (19.4)	0.788
Vaginal delivery with prior CS	4 (9.8)	2 (6.5)	0.693
Cesarean delivery	32 (78.0)	25 (80.6)	0.788
Cesarean delivery with prior CS	9 (22.0)	9 (29.0)	0.492
Pregnancy complications
Preeclampsia	7 (17.1)	3 (9.7)	0.579
Gestational diabetes mellitus	7 (17.1)	3 (9.7)	0.579
Macrosomia	4 (9.8)	1 (3.2)	0.541
Multiple pregnancy	5 (12.2)	1 (3.2)	0.351
Uterine myoma	5 (12.2)	6 (19.4)	0.613
Placenta previa	15 (36.6)	6 (19.4)	0.111
Placenta accreta	12 (29.3)	2 (6.5)	0.015

Data are presented as mean ± standard deviation or n (%).

CS, cesarean section.

Bold P values are statistically significant.

Comparison of emergency exploratory laparotomy circumstances

In the hysterectomy group, 34 (82.9%) patients had vaginal hemorrhage as the primary abnormal sign after returning to the ward. In the uterine preservation group, 15 (48.4%) patients had vaginal hemorrhage and 14 (45.2%) patients had a pelvic hematoma as the primary abnormal signs. There were no statistically significant differences in the operation time between the two groups. The amounts of hemorrhage (P = 0.010), erythrocyte infusion (P < 0.001), frozen plasma utilization (P = 0.030), apheresis platelet suspension (P = 0.004), and cold precipitation (P =0.013) in the hysterectomy group were greater than those in uterine preservation group, while the frequency of uterine artery ligation was lower in the hysterectomy group than in the uterine preservation group (P < 0.001) (Table 2).

Table 2.

Circumstances of emergency exploratory laparotomy.

	Hysterectomy (n = 41)	Uterine preservation (n = 31)	P
Abnormal signs after returning to the ward
Vaginal bleeding	34 (82.9)	15 (48.4)	0.002
Pelvic hematoma	6 (14.7)	14 (45.2)	0.004
Persistent fever	1 (2.4)	2 (6.4)	0.574
Operating time, minutes	138 ± 61.28	121 ± 66.94	0.269
Measures to prevent hysterectomy
Ligation of large vessels	5 (12.2)	16 (51.6)	<0.001
Tamponade of the uterus	9 (22.0)	2 (6.5)	0.139
B-Lynch procedure	4 (9.8)	9 (29.0)	0.035
Bakri balloon	9 (22.0)	3 (9.7)	0.166
Blood loss, mL
Blood loss before laparotomy	2895 (1000–15000)	1500 (400–4190)	0.010
Total blood loss	4983 (1800–16000)	2170 (995–6000)	0.010
Transfusion blood products
Erythrocyte suspension	18 (2–80)	4 (2–20)	<0.001
Fresh frozen plasma	1800 (400–8500)	800 (400–2600)	0.030
Apheresis platelet suspension	1 (0–4)	0 (0–1)	0.004
Cryoprecipitate	2 (0–48)	0 (0–10)	0.013
Fibrinogen derivatives	4 (0–24)	1 (0–20)	0.352
Human albumin	50 (0–190)	20 (0–190)	0.810

Data are presented as n (%), mean ± standard deviation, or median (range).

Bold P values are statistically significant.

Indications for emergency laparotomy

All 72 patients undergoing emergency exploratory laparotomy were grouped according to the initial indications for this operation: uterine inertia (31 patients, 43.0%), placental factors (27 patients, 37.5%), coagulation dysfunction (9 patients, 12.5%), and uterine rupture (5 patients, 6.9%). The frequencies of placental factors and coagulation dysfunction were greater in the hysterectomy group than uterine preservation group, while the frequency of uterine inertia was greater in the uterine preservation group than hysterectomy group. There were no statistically significant differences in the frequency of uterine rupture (Table 3).

Table 3.

Main reasons for exploratory laparotomy.

	Hysterectomy(n = 41)	Uterine preservation(n = 31)	P
Uterine atony	13 (31.7)	18 (58.1)	0.025
Placenta
Placenta accreta	12 (29.3)	2 (6.5)	0.015
Placenta previa	7 (17.1)	6 (19.4)	0.803
Coagulation disorders
Amniotic fluid embolism	4 (9.8)	1 (3.2)	0.382
Acute fatty liver of pregnancy	1 (2.4)	0 (0.0)	1.0
HELLP syndrome	2 (4.9)	1 (3.2)	1.0
Uterine rupture	2 (4.9)	3 (9.7)	0.646

Data are presented as n (%).

Bold P values are statistically significant.

Comparison of hematological and coagulation factors between groups

Both the final preoperative and first postoperative hematological and coagulation-related laboratory parameters for patients undergoing exploratory laparotomy are compiled in Table 4. The preoperative prothrombin activity (P = 0.001), fibrinogen level (P = 0.001), and platelet count (P = 0.005) were significantly lower in the hysterectomy than uterine preservation group, while the preoperative international normalized ratio (P = 0.001) and activated partial thromboplastin time (P = 0.019) were significantly greater in the hysterectomy than uterine preservation group. Among the postoperative parameters, there was no significant difference in the postoperative international normalized ratio, activated partial thromboplastin time, or fibrinogen level. However, the platelet count (P = 0.002) and prothrombin activity (P = 0.048) were significantly different between the hysterectomy and uterine preservation groups.

Table 4.

Changes in blood counts associated with exploratory laparotomy.

	Hysterectomy(n = 41)	Uterine preservation(n = 31)	P
Prothrombin time activity
Preoperative	49.48 ± 26.52	79.57 ± 34.20	0.001
Postoperative	64.98 ± 17.86	78.35 ± 28.66	0.048
INR
Preoperative	2.20 ± 1.18	1.25 ± 0.33	0.001
Postoperative	1.42 ± 0.37	1.23 ± 0.26	0.06
APTT, s
Preoperative	61.42 ± 42.27	42.8 ± 20.62	0.019
Postoperative	44.9 ± 24.6	35.7 ± 16.19	0.069
Fibrinogen, mg/dL
Preoperative	1.54 ± 1.14	2.5 ± 1.23	0.001
Postoperative	2.4 ± 1.26	2.61 ± 1.19	0.66
Hemoglobin, g/dL
Preoperative	69.97 ± 23.7	73.6 ± 23.1	0.31
Postoperative	84.92 ± 21.73	85.7 ± 12.55	0.84
Platelet count, K/µL
Preoperative	124.24 ± 85.99	182.85 ± 76.72	0.005
Postoperative	84.75 ± 51.93	130.74 ± 61.55	0.002

Data are presented as mean ± standard deviation.

INR, international normalized ratio; APTT, activated partial thromboplastin time.

Bold P values are statistically significant.

Comparison of maternal complications

Among all 72 patients, 44 (61.1%) had complications, including secondary disseminated intravascular coagulation (DIC) in 33 patients, traumatic infection in 7 (9.7%), urinary system injury in 4 (5.6%), acute renal insufficiency in 2 (2.8%), pneumonia in 2 (2.8%), cardiac ischemia in 2 (2.8%), and death in 1 (1.4%). The difference in the maternal hospitalization time between the hysterectomy and uterine preservation groups was not statistically significant (Table 5).

Table 5.

Complications and fetal outcomes of exploratory laparotomy.

	Hysterectomy (n = 41)	Uterine preservation (n = 31)	P
Total complications	29 (70.7)	15 (48.4)	0.054
Disseminated intravascular coagulation	23 (56.1)	10 (32.3)	0.044
Wound infection	4 (9.8)	3 (9.7)	1.0
Urinary system injury	2 (4.8)	2 (6.5)	1.0
Acute renal insufficiency	2 (4.8)	0 (0.0)	0.503
Pneumonia	2 (4.8)	0 (0.0)	0.503
Cardiac ischemia	2 (4.8)	0 (0.0)	0.503
Death	1 (2.4)	0 (0.0)	0.569
Postoperative hospital stay, days	11 (5–38)	9 (4–32)	0.375

Data are presented as n (%) or median (range).

Bold P values are statistically significant.

Discussion

Few clinical reports to date have focused on emergency exploratory laparotomy. In one study, the primary reason for emergency exploratory laparotomy in the obstetrics and gynecology department was severe or persistent postoperative bleeding.[9] In the present study, the primary abnormal signs exhibited by patients returning to the ward were vaginal hemorrhage and persistent or excessive postpartum hemorrhage, resulting in different degrees of hemorrhagic shock as well as coagulation dysfunction in some cases. Ineffective conservative treatment and immediate danger to the patient’s life are the main surgical indications for emergency exploratory laparotomy. To determine whether such an operation needs to be performed, the patient’s vital signs and the presence of uterine atony must be closely monitored after returning to the ward. If vaginal bleeding, abdominal bulging, shifting dullness, or other abnormal symptoms are present, relevant auxiliary examinations should be rapidly implemented to quickly establish the cause of bleeding. When conservative treatments fail to control these symptoms, a hemostatic operation or emergency hysterectomy is often essential to preserve the life of the mother. After undergoing a hysterectomy, however, women of reproductive age may experience both physical and psychological discomfort, leading to a reduction in quality of life. Therefore, the main risk factors leading to hysterectomy must be more thoroughly explored and understood.

In recent years, placental factors have replaced uterine inertia as the most common indications for emergency hysterectomy.[10,11] The primary indications for hysterectomy in the present study were placental factors, coagulation disorders, uterine inertia, and laceration of the birth canal. Traumatic endometrial injuries or endometrial dysplasia resulting from a history of multiple operations can result in pathological placental adhesion characterized by incomplete detachment of the placental and fetal membranes after delivery, leading to substantial bleeding. Since implementation of the second child policy in China, the number of patients with dangerous placenta previa or placental implantation has substantially increased because of the high rates of early cesarean section, and these patients have a significantly higher risk of intraperitoneal hemorrhage than those who undergo typical cesarean section.[12,13] A uterine hemorrhage surge occurs during anterior placenta cesarean section. When this is accompanied by implantation that penetrates the placenta, the local tissue can become difficult to identify, resulting in inaccurate intraoperative hemostasis. In other cases, patients with hemorrhage may develop shock due to a low blood volume, with vasoconstriction and low blood pressure resulting in a lack of clear hemorrhage within tissues.[14,15] Once this initial shock has been corrected and the blood volume has been restored, the spasmodic blood vessels can reopen, resulting in reactivation of bleeding that necessitates a secondary surgery. Karaman et al.[16] demonstrated that local resection might be a promising conservative cesarean hysterectomy technique for reduction of postpartum bleeding in the management of placenta percreta. These findings suggest that different uterine perseveration techniques might also result in different outcomes. Thus, it is important to distinguish the etiology of and choose an appropriate treatment for placenta percreta.

Substantial postpartum hemorrhage can result in the loss of large volumes of coagulation factors and platelets, leading to coagulation dysfunction.[17,18] In the present study, we found a high frequency of uterine removal as a result of such coagulation disorders; seven patients with such disorders underwent emergency exploratory laparotomy, of whom five ultimately underwent hysterectomy. Extensive blood loss as a result of postpartum hemorrhage can result in hemorrhagic shock or coagulation dysfunction followed by microvascular embolization and multiple organ failure, resulting in a high mortality rate.[19,20] We found that the preoperative prothrombin activity, fibrinogen level, and platelet count were significantly lower and that the international normalized ratio and activated partial thromboplastin time were significantly higher in the hysterectomy group than in the uterine preservation group, suggesting a higher degree of coagulation dysfunction in patients who ultimately underwent a hysterectomy. Zhang et al.[21] indicated that a postpartum coagulation enzyme activity level of <50% was the greatest risk factor for hysterectomy in their study of perinatal total hysterectomy. This finding suggests that severe uterine bleeding due to dysfunctional coagulation can be avoided or reduced by actively seeking to improve relevant coagulation indexes, continuously monitoring hemodynamic changes, and preventing the occurrence of coagulation disorders. Timely transfusion with appropriate blood components is essential for the patient’s well-being. A massive transfusion protocol such as that used at the Stanford University Medical Center (6 U of red blood cells, 4 U of plasma, and 1 U of platelet composition to comprise a 6:4:1 unit for transfusion) or repeated transfusion based on the amount of bleeding can be administered. During the later stages of patient resuscitation, a prolonged prothrombin time, decreased platelet count, and low baseline fibrin level can result in the need for increased plasma, fibrinogen, or cold precipitation infusion to prevent DIC and indirectly prevent uterine resection. In the present study, the incidence of complicated DIC was significantly greater in the hysterectomy group (56.1%) than in the uterine preservation group (32.3%). When primary or secondary coagulation mechanism disorders are not corrected in a timely and effective manner, postoperative bleeding or abdominal hemorrhage can readily occur. Comprehensive assessment of the bleeding volume, bleeding rate, early rapid rehydration, timely blood transfusion, and prevention of DIC are all essential to improve patient outcomes.

Uterine inertia is the most common cause of postpartum hemorrhage.[22,23] We found that the incidence of uterine atony was high but that the proportion of uterine resection was low in the present study. New-onset strong uterine atony, which is an indication for the administration of prostaglandin preparations, is associated with higher rates of successful conservative treatment in patients with postpartum hemorrhage caused by uterine inertia. Postpartum hemorrhage can be treated in a variety of ways, including uterine packing, uterine compression using a B-Lynch suture, pelvic vascular ligation, abdominal aorta or internal iliac artery occlusion, or pelvic artery interventional embolism. However, the exact conditions in patients with postpartum hemorrhage are complex and frequently change. Surgical treatment should thus be individualized to select the fastest, simplest, most skilled, and least traumatic mode of hemostatic operation. If a variety of hemostatic measures have been used in the first operation and patients still present with severe uterine bleeding, then cesarean section and hysterectomy may reduce the need for a secondary exploratory laparotomy. Further efforts to preserve reproductive functionality and to better understand the circumstances necessitating hysterectomy will be invaluable in the future.