Management
Following uterine inversion, prompt treatment of hemorrhage and shock is vital in limiting maternal morbidity and the risk of mortality. Hypotension and hypovolemia require aggressive fluid resuscitation. The general principles of treatment follow the STAR protocol (ie, shock, treat aggressively, and repair). The best treatment of acute puerperal inversion is immediate uterine repositioning. The approach to treatment should follow a logical progression.
Shock
Summon help.
Initiate fluid resuscitation with 2 large-bore intravenous lines. Promptly administer 1 or more liters of an isotonic salt solution such as lactated Ringer parenterally. Hetastarch also may be infused, as required, to support arterial pressure or restore circulating volume.
Submit specimens to the laboratory for possible transfusion and for determination of baseline values of hemoglobin (Hgb), hematocrit (Hct), and coagulation factors.
Insert a Foley catheter.
Immediately summon an anesthesiologist.
Treat aggressively
Order appropriate surgical equipment and summon assistants to ready the operating room for a possible laparotomy.
Administer tocolytics to promote uterine relaxation. These may include nitroglycerin 0.250-0.500 mg IV over 1-2 minutes, terbutaline 0.100-0.250 mg IV slowly, or magnesium sulfate at 4-6 g IV over 20 minutes.
Attempt prompt uterine replacement. First, proceed with a trial of simple manual replacement. If this is unsuccessful, promptly perform a laparotomy for a surgical replacement unless the clinician is trained in one of the vaginal approaches. At laparotomy, general anesthesia employing a uterine relaxing agent is best, especially if a tocolytic has not been given or if the drug administered does not appear to work.
Repair
Suture birth canal lacerations and any surgical incisions in cervix or vagina.
Perform uterine massage (after replacement).
Administer uterotonics. These may include methyl ergonovine maleate (Methergine 0.2 mg IM q30min 3 times prn), oxytocin by constant infusion (40-60 IU/L in an isotonic salt solution), prostaglandin 15-methyl F2 alpha (Hemabate 0.25 mg IM q30min 3 times prn), or misoprostol (0.4 mg PO or buccal q2h prn, alternatively 0.8-1 mg PR once).
Monitor closely after replacement of the uterus to avoid reinversion.
The best technique for rapid uterine replacement is controversial. Theoretically, manipulating the genital organs with restretching of the pelvic viscera in the presence of some degree of neurologic shock could potentiate the problem. However, as a practical matter this risk is insignificant. Based on long clinical experience, the best management is always immediate manual replacement. The longer the delay, the greater the risk for complications of cardiovascular collapse and its serious sequelae. Also, the longer the delay, the more contracted the lower uterine segment and/or cervix becomes, rendering the replacement progressively more difficult.
Because most deliveries now take place in labor rooms, this site is probably the most frequent venue for trials of immediate replacement if the placenta has separated. However, depending on the circumstances for whether the placenta remains attached, a reasonable delay may be prudent while help (eg, anesthesiologist, nurses, assistants) is recruited, laboratory tests are sent, blood bank assistance is assured, aggressive fluid resuscitation is initiated (with a balanced salt solution such as lactated Ringer), and the parturient is transported to an operating room. In the operating room, reduction of the uterine inversion may be accomplished either by a conservative, nonsurgical approach or via a surgical procedure as described below.
Nonsurgical techniques
The literature includes descriptions of a number of nonsurgical replacement techniques. In the past, intravaginal packing was advocated, as was the use of a pessary. These procedures are now of historical interest only.
In 1945, O'Sullivan described a method for the correction of partial inversion using simple hydrostatic pressure. In this technique, warm sterile water or isotonic sodium chloride solution is rapidly instilled into the vagina via a tube while the introitus is blocked either with an instrument such as a vacuum extractor or the accoucheur's forearm. The instilled fluid progressively distends the vaginal wall and then forces the fundus upward to restore it to its original position. Some clinicians favor a trial of this procedure in selected cases because of its simplicity.
With the advent of potent tocolytics, the technique of manual replacement has been greatly simplified. A variant of the manipulations originally described by Johnson in 1949 is recommended. In this procedure, the operator's hand is placed in the vagina, cupping the inverted fundus in the palm. The uterus is then firmly and promptly lifted upward in the pelvic curve through the pelvis and into the abdominal cavity to the level of the umbilicus. This manipulation forces the uterine ligaments to stretch. Usually, when the inverted mass is pushed upward, the uterus promptly reverts, with the fundus returning to its anatomic position. If the reversion is successful, the uterus is held in place for several minutes. Uterotonics are then administered to firm the myometrium.
To facilitate the uterine replacement, terbutaline, magnesium sulfate, and nitroglycerin have been successfully employed as uterine relaxants, with and without concomitant general anesthesia. In the immediate period following successful replacement, serial doses of uterotonics are administered to avoid reinversion and avoid secondary atony and hemorrhage.
Once the correct diagnosis is made, the surgeon should act with celerity before the myometrium regains its tone. Under these circumstances, uterine replacement may be surprisingly easy and use of a tocolytic and various complex manipulations to return the uterus to its correct anatomic position may be avoided. As discussed below, a potential exception to this rule of immediate action is inversion with the placenta remaining attached to the uterus.
If the placenta has not separated before the replacement operation is attempted, the most prudent approach is to leave it undisturbed until the patient is in the operating room and an anesthetic has been administered or a tocolytic given. Immediate placental removal without a successful replacement simply increases blood loss. Further, in the uncommon event of a placenta accreta, increta, or percreta, removal proves either difficult or impossible. In this setting, removal efforts markedly increase the blood loss and waste valuable time. Unfortunately, the degree of abnormal placental adherence cannot be established until removal is attempted and then it may be too late. Therefore, if inversion occurs and the placenta is still attached, the best plan is to wait for the safer environment of the operating room rather than to attempt an immediate and perhaps incomplete placental removal in another, less well-equipped setting.
Exceptions to this general rule may be necessary in certainclinical settings where equipment or surgical assistance is limited.
Surgical techniques
If 2 or more attempts at manual replacement are unsuccessful, surgery is indicated. An abdominal approach for uterine replacement is favored. A vaginal technique has also been described but has few adherents.
In the vaginal procedure, the bladder is dissected from the cervix, and the anterior lip of the cervix and the anterior wall of the uterus are incised to the extent necessary to permit replacement. After the uterus is repositioned, the uterine wall and cervical defects are repaired in layers. This operation is not recommended except for those especially trained in its performance.
The favored transabdominal technique is a modification of the procedure originally described by Huntington in 1921. Deep general anesthesia is administered with agents promoting uterine relaxation. A laparotomy is performed, and the round ligaments identified. The surgeon then grasps the round ligaments with an atraumatic clamp such as a Babcock and applies gentle upward traction. The operator pulls each round ligament up into the peritoneal cavity. The round ligaments are then released and immediately grasped again more distally, akin to pulling on a rope, hand over hand. This maneuver is repeated until the fundus is completely restored to it normal configuration. If available, a second operator, applying upward pressure from below, facilitates the procedure. As the uterus begins to revert, the lower segment is squeezed like a tube of toothpaste to accelerate the process. As in the manual replacement technique, uterotonics are administered as soon as the uterus has returned to its normal shape.
In the unusual instance that the Huntington operation is unsuccessful, the more extensive Haultain procedure is required. In this technique, a posterior, longitudinal hysterotomy incision is performed. This acutely widens the lower uterine segment and facilitates uterine replacement. The posterior uterine wall is incised to avoid inadvertent injury to the bladder. Thereafter, upward traction on the round ligaments follows, as described above for the Huntington procedure, to complete the uterine reversion. The defect is repaired in layers.
Regardless of the procedure employed, after repositioning, immediate uterine atony is common and prompt reinversion may occur. Administration of 15-methyl F2 alpha prostaglandin (Hemabate), high-dose oxytocin, methyl ergonovine maleate (Methergine) parenterally, or misoprostol per rectum is recommended. If magnesium sulfate was administered as a tocolytic, calcium gluconate can be administered to reverse the tocolytic effect.
Administration of antibiotics has been advocated when inversion occurs, based on the theory that the various manipulations for replacement predispose to infection. This risk is probably small, albeit unknown. Many clinicians do administer prophylactic doses of a broad-spectrum first-generation cephalosporin or a similar drug following repositioning. There are no data concerning the necessity of such treatment.
Comments
The basic pathophysiologic mechanism for inversion is not known. Nonetheless, several observations of clinical cases of inversion as well as various experiments in nonhumans are worth review. As previously mentioned, inversion does rarely occur at cesarean delivery. This may happen spontaneously or following administration of a potent tocolytic such as nitroglycerin. In these and other cases of acute inversion, the flaccidity of the myometrium is striking. If the uterus remains flaccid in the moments immediately following parturition and the placenta is fundally implanted, then a downward protrusion of the fundus is possible and probably quite frequent. Under unique circumstances this myometrial flaccidity (perhaps aided at this critical juncture in some instances by the classic inappropriate cord traction) permits the fundus to indent, with or without the placenta remaining implanted.
To result in an inversion, the uterus must next resume contracting at precisely the right moment and force the inverted fundus or fundus and placental mass downward, driving it deeper into the lower uterine segment. If the cervix is sufficiently dilated, the mass can be squeezed through this barrier, resulting in a complete inversion. In less extreme situations, the indented fundal wall is itself trapped within the middle of the uterine cavity, resulting in a partial inversion. In complete inversions, the cervix functions as a constricting band once the fundus has passed through it and edema rapidly forms. The prolapsed mass then becomes progressively larger, more completely obstructing venous and arterial flow. Serious tissue injury or necrosis is a possible, albeit rare, sequela in chronic or neglected cases.
The importance of uterine tonus in the etiology of inversion is emphasized by the experience with the protocol for the active management of the 3rd stage of labor (AML). When the data from a large clinical trial of AML were analyzed, an unanticipated benefit was that the incidence of uterine inversion was markedly reduced in comparison to the control group. This occurred despite the routine use of immediate cord traction following delivery of the infant, part of the AML protocol. The important difference, however, in the AML protocol versus natural birth is the routine administration of intravenous oxytocin immediately following the delivery of the neonate but before signs of placental separation. Based on these observations, maintenance of uterine tonus appears to be critical to preventing an inversion.
When considering the pathophysiology of inversion, one should wonder why it does not happen with greater frequency. Postpartum uterine flaccidity, active traction on the cord, and fundal implantation of the placenta are frequent; however, inversion remains, at best, a rara avis.
Inversion is rarely observed in nonpregnant individuals. The characteristic finding in these cases is the presence of a pedunculated subserosal leiomyoma or other neoplasm, located at or adjacent to the uterine fundus. Most unusually, endometrial polyps are the culprit. In these gynecologic cases, the mechanism for inversion is presumed to be much as it is in the puerperium, although the time interval for the inversion is likely much longer. The tumor or mass descends into the lower segment of the uterus. The uterus then responds with recurrent contractions, driving the mass lower and, on occasion, entirely through the cervix.
In the rare case when a pregnancy-related inversion is long-standing or if a manual replacement is unsuccessful and surgical replacement is required, consultation with an experienced clinician, if possible, is advised. At surgery, an above-and-below procedure with 2 operators speeds the procedure and reduces the risk of iatrogenic injury.
The recurrence risk of inversion in subsequent deliveries is unknown. Similarly, the morbidity rate from inversion cannot be calculated with accuracy. However, maternal deaths do occur in association with inversion. Fully 90% of these happen within 2 hours or less after delivery. These facts emphasize the importance of celerity on the part of the birth attendants to replace the uterus and aggressively support the mother once the diagnosis is established.
Uterine inversion is a relatively frequent problem in veterinary medicine and is particularly dangerous and difficult to resolve among bovines. A chance observation in mice hints at the possibility of a genetic propensity for inversion that might have implications for human medicine. In specially bred mice, females homozygous for a gene locus that codes for deficiencies in melanocytes, mast cells, and osteoclasts have a remarkably high incidence of postpartum uterine inversion (86%). In some of these cases, the uterus was observed to invert with the placenta still implanted.
Following a series of subsequent cross-mating studies, the maternal genotype (not that of either the father or the fetus) was determined to be the important component in coding for this inversion risk. This unusual experiment hints strongly that specific genes exist that code for abnormal uterine tone during the immediate puerperium. This suspected functional disorder of the myometrium predisposes to puerperal inversion by an unknown mechanism. These data provide a new and fascinating hypothesis for future investigation.