Management of Acute and Subacute Puerperal Inversion of the Uterus PETER WATSON, MD, NICHOLAS BESCH, MD, AND WATSON A. BOWES, Jr, MD Eighteen cases of acute and subacute puerperal inversion were studied to identify important predisposing factors and to assess management and postpartum morbidity. The study patients did not differ from case-matched controls in age, party, duration of labor, type of delivery, or manage- ‘ment of the third stage. The most common signs noted ‘were hemorrhage (94%) and shock (39%). All inversions were recognized immediately and manually replaced within 60 minutes, Shock was treated prior to repositioning in all cases, Calculated blood loss averaged 1775 ml. There ‘was no mortality nor febrile morbidity. The average hospi- lal stay of the study patients and their case-matched con- trols was 3 days. Immediate recognition of uterine in- version and prompt initiation of therapy will ensure a normal postpartum course. Acute and subacute puerperal inversion of the uterus constitutes an obstetric emergency with significant mortality, despite modern availability of blood trans- fasion and antibiotics.' The experience of the authors over the previous 7 years was reviewed and 18 cases of acute and subacute uterine inversion were noted, With the use of case-matched controls, the authors at- tempted to identify important predisposing factors and significant differences in postpartum morbidity Puerperal inversion of the uterus is classified as acute if the inversion has occurred without cervical contraction, subacute if cervical contraction has oc- curred, and chronic if more than 4 weeks have elapsed since inversion and cervical contraction.» The version is termed incomplete if no part of the corpus inverts past the cervix, complete if there is inversion beyond the cervix, and prolapsed ifthe inverted uterus ‘extends beyond the introitus.** Tom the Deparment of Obsttrics and Grecia, Uniorsily of (Colorado Metical Contr, Denver, Colorado. ‘Submited or pubieation Ap, 1979, 12 0029-7844/80/010012-05$01.75. Materials and Methods Between January 1, 1969, and January 1, 1976, 31,299 deliveries occurred at the Denver General and Colo- rado General Hospitals. During this time, there was a total of 18 uterine inversions, an incidence of 1 in 1739 deliveries, There were 8 acute inversions and 10 sub- acute inversions. All eases involved complete in- version with prolapse beyond the introitus ‘These 18 cases were analyzed for predisposing fac- tors, management of the inversion, and postpartum morbidity and were then compared to 2 groups of ‘case-matched controls. Control group A was formed by selecting the delivery that occurred immediately following each inversion, Control group B was formed by selecting the next delivery after those in group A, of equal parity to the matched inversion, Net blood loss was calculated by multiplying the percentage of hematocrit change from admission to discharge by 150 il per percent and adding the amount of blood trans- fused. Differences between the study patients and the control groups were analyzed for significance by x° and the Student's f tests Results The results of the present study are summarized in Tables 1 and 2, The mean age of the study patients ‘was 22 years with a range of 16 to 33, The mean parity was 2. The mean infant weight was 3269 g with a range of 2340 to 4370 g. There was no significant dif- ference in age, infant weight, or parity between the pa- tients and their case-matched controls, ‘The mean durations for the various stages of labor were 387 minutes, 40 minutes, and 17 minutes for the first, second, and third stages, respectively. The mean. third stage duration in the study group was signifi- cantly greater (P < 0.05) than in control group A, but it was not significantly different when controlled for parity (control group B). Three of the 18 labors re- VOL. 55, NO. 1, JANUARY 1980 Obstetrics & GynecologyTable 1. Uterine Inversion Study Group: Clinical Characteristics Infant Anesthesia at weight Managementof Location of Patent ‘Age Pasty Typeof delivery delivery (@ the placenta the placenta mH 273 Spontaneous vertex Padendall 2300 ‘Traction Fundal TT 1ST __Spontaneous vertex Pudendal 20 Cree Fondal AR 201 Spontaneous vertex Pudendal 2600 Traction Fundal 1H 25-2 Spontaneous vertex NO, local 3540 Traction Fundal vs 9 1 Assated breceh Pudendal 3240 Manual removal Anterior, fndal tc 222 Spontancous vertex Pudendal 3230 Traction Lateral, funda FG 233 Spontancous vertex” Padendal 2700 Traction Fundal ™ 291 —_Low forcepe vertex Saddleblock 3160 Traction Unknown BR 201 Spontaneous vertex Epidural 3850 Traction Fandal DDN 202 Spontaneous vertex Padendal ‘60 Traction Unknown DUD 191 Mid forceps vertex Epidral 4370 Trsetion Unknown bw 33 5—_Spontaneous vertex Pudendal 3080 Traction Unknown sk 17 1 Low forcepe vertex Pudendal 3460 Traction Unknown rR 212 Spontaneous vertex” Pudendal 2560 Traction Fundal na 22 2 Spontancous vertex None 220 Traction Fundal MP 27 2—_Spontaneous vertex Pudendal 3740 Traction Unknown ev 23 2——_Law forceps vertex Saddleblock 2460 Traation Fundal Do 222 Spontaneous vertex" DemeroliV 3450. Traction Fandal Men 222 3269 Oxytocin induced or augmented quired oxytocin for augmentation and/or induction, as did 2 labors in each control group. All of the patients delivered vaginally. There was 1 assisted breech vaginal delivery and 4 forceps deliv- cries in the study group. Each delivery in the control groups was a vaginal vertex delivery, with 4 forceps deliveries in control group A and 6 forceps deliveries, in control group B. Regional anesthesia, either by con- tinuous lumbar epidural or by subarachnoid block, ‘was administered for delivery in 4 of the 18 cases of inversion. All other deliveries were performed under either pudendal or local anesthesia, There was no sig- nificant difference in the incidence of delivery under regional anesthesia in either control group. ‘Table 2. Uterine Inversion Study Group: Additional Clinical Characteristics Duration of labor (min) Durationof—_Caleulated Volume . first/second/ inversion net blood transfused Anesthesia for Hospital Patient thie stages (evi) loss (ml) Shock (el) repositioning stay (lays) RH 150/30/25, 2s 3500 + 2000 Halothane/NO 6 1 2180/15/40 10 1500 S © Halothane/N-O 3 an 500/15/18 BR >1s0 - © Pudendal 3 ri 210/18/25 9 1500 + © Halothane/N.O 4 vs 1630/27/83 ° 1600 + 1000 Halothane/N,O 3 uc 30/30/18 5 1680 - © Pudendal 2 FG 195/2/13 « 3350 + 2000 —_Halothane/N.O- 4 mM 495/512 7 1050 = © Halothane/N,O 3 BR 765/38/10 8 1500 - © Hlothane/N.O 4 DON 317/6/9 © 2900 + 2000 _-Halothane/NO- 4 DLD '10/206/4 2 2200 1000 Halothane /NO- 4 bw 590/15/10 a 2650 + 2500 Cyclopropane 4 SR 310/65/10 1 780 - 0 Pudendal as FR 300/21/32 2 1050 - 0 Pudendal 2 1280/20/15 4 1350 © None 3 ME 1705/24/27 1 2950 1000 —Padendal 3 EV 150/58/22 10 1100 a 500 Saddle block 3 bo 195/28/9 1 1800 - © Demerol IV 2 Mean 387/40/17 (ain) 8 175 34 * Calculated blood los VOL. 5S, NO. 1, JANUARY 1980 (soz 189(admieion hematoct~ dscharge hematocri + volume transfused Watson etal Uterine hnversion 13,In each of the study cases, the placenta was deliv- ered at least partially attached to the inverted uterus. ‘Only 1 case was recorded of attempted manual re- moval of the placenta prior to inversion and only 1 ‘case of placental delivery by a Crédé maneuver. In all other cases, the placenta was delivered by cord trac- tion with the uterus stabilized above the pubic sym- physis (Brandt technique). A careful review of the rec- ords revealed no evidence of excessive traction on the cord, In each of the 2.control groups, there was 1 man- ual removal of the placenta, All other placentas were delivered by cord traction as described above. Each inversion was managed by manual replace- ment of the uterus after the placenta had been sepa- rated from the uterine wal. In the 10 subacute cases, general anesthesia was required to obtain adequate re- Taxation of the cervicouterine junction. In each case, manual replacement was accomplished by broad liga- ment counter traction. As originally described by Johnson, this method involves grasping the fundus in the paim of the hand with the fingers extended to the posterior fornix. The uterus is then lifted toward the umbilicus and gradually repositioned by steady, mod- crate pressure (Figures 1-3) In each case, the uterus was manually explored after replacement. No uterine rupture or evidence of pla- centa acereta was found on exploration. In 12 of the 18 cases, the placenta was thought to have been im- planted on the fundus. Inthe remaining 6 cases, no es- timation was made as to the placental site. Antepar- ‘tum ultrasonography in 2 study patients demonstrated fundal implantation of the placenta. In all cases fol- lowing replacement and exploration, intravenous oxytocin (usually 30 U/1000 ml fluid), 0.2 mg methyl- ergonovine (intramuscularly), and occasionally cal- cium gluconate (usually 44 mEq/liter) were given to promote uterine contraction and hemostasis. None of Figure 1. The invered fundus grasped in the palm of the hand with the fingers dizccted toward the posterior fornix The placenta i Toft attached if possible 14 Watson etal Uterine Inversion Figure 2. The uterus lied out of the pelvis and directed with steady pressure toward the umbilicus, these agents was ever administered during the third stage of labor prior to the delivery of the placenta. Shock was noted in 7 of the 18 study patients, and in each case it was immediately treated by volume re- placement and pressor agents when necessary. No shock was seen with any of the controls (P < 0,001). In zo instance was the degree of shock thought to be out of proportion to blood loss. The patients were trans- fused when necessary and a hematocrit obtained just prior to discharge from the hospital. The mean calcu- lated blood loss of 1775 ml was significantly greater than the mean blood loss for either control group (P < 0.001). The blood loss was directly related to the dura- tion of the inversion (r = 0.8) except in patient MF, in whom an untreated partial placental separation re- sulted in a hemorrhage of approximately 2000 ml prior to inversion. No patient developed postpartum hypopituitary function. Several of the patients were given medication for lactation suppression, but exami nation of their subsequent hospital records revealed no evidence of hypopit Figure 3._"The repositioned uterus prior to exploration and manual removal of the placenta Obstetrics & Gynecology