EMERGENCY HYSTERECTOMY : The Hobsonís Choice in MassiveAtonic Post Partum Haemorrhage
Sanjay B Rao, PK Shah, S Dholakia, N Shreyan, N Vanjara
*Lecturer; **Professor, Unit Head; ***Senior Registrar, Department of Gynaecology, LTMG Hospital, Sion, Mumbai.
This retrospective study in a tertiary institute analyses 58 cases of uncontrolled exclusive atonic postpartum haemorrhage excluding traumatic causes where emergency hysterectomy was performed as the last resort. Out of 48252 deliveries during the period from Jan., 1990 to June, 1999 there were 58 such patients. Amongst patients undergoing obstetric hysterectomy, 78.7% had atonic haemorrhage. 56.8% were between 20-30 years; 67.2% were multigravidae. 72.4% were unregistered throughout pregnancy. Anaemia, mismanaged IIIrd stage, injudicious use of oxytocics and antepartum haemorrhage were the leading risk factors identifiable in the study group. 84.4% patients underwent a subtotal hysterectomy and 18.9% patients also had a hypogastric artery ligation.
Eastman, in 1950, had stated that in modern obstetrics, no woman should die from post-partum haemorrhage. Unfortunately, it would be appalling to note that even today, 8-10% of maternal mortalities in developing countries directly occur due to massive obstetrical haemorrhage. There is no reliable data on the true incidence of severe life threatening postpartum haemorrhage. Maternal morbidity and mortality rise not only with delay in diagnosis but also in accordance with any increase in the caesarean section rate. Though traditionally defined as blood loss in excess of 500 ml during the first 24 hours after delivery, there is no specific universally accepted definition of postpartum bleeding. The core problem is that estimation of blood loss during delivery is notoriously inadequate because it is based on the visual observation of the obstetrician rather than any objective measurement. In a recent and more universally acceptable study, a haematocrit change of 10% or a need for red cell transfusion was adopted as a definition of post partum bleeding. 
The atonic uterus could lead to a catastrophic bleeding. The main aetiological factors could be retained placental bits, extreme, uterine distension leading to disruption of actin myosin filaments; precipitate labour or uterine hypoxia causing muscle exhaustion due to lactate build up or glycogen depletion.
MATERIAL AND METHOD
This study was conducted at a 1400 bedded multidisciplinary tertiary referral teaching institute. The main aim was to closely evaluate 58 cases of exclusive primary atonic postpartum haemorrhage not responding to aggressive conservative management. Medical management included use of oxytocics, volume replacement, bimanual massage of uterus, exclusion of traumatic causes and coagulopathy. Tight intrauterine pack was done in two cases. Emergency hysterectomy was performed as the last life saving resort. Each of these 58 cases were analysed in terms of age, parity, associated high risk antenatal, intranatal and postnatal factors and the mode of delivery. Foeto-maternal outcome was assessed in terms of morbidity and mortality.
There were 48, 252 deliveries during the period from Jan., 1990 to June, 1999. Out of 74 (0.15%) cases of obstetric hysterectomy during this period, there were 58 cases of exclusively atonic postpartum haemorrhage. Hence the incidence of atonic postpartum haemorrhage was 78.7% amongst patients undergoing emergency hysterectomy. All hysterectomies were performed by either lecturer, associate professor or unit head. Majority of these cases were in the age group 20-30 years (56.89%) as shown in Table 1. There were 12 grand multiparae in the study group (20.68%). Seven were primigravidae (12.06%) and 39 multigravidae (67.24%) (Table 2). An important risk factor identifiable in the study series was that a vast majority of 42 cases (72.41%) were unregistered throughout pregnancy. They presented as emergency admissions directly in labour. 16 (27.5%) patients had antenatally registered at this institute but had less than two antenatal visits. Thirty eight cases out of 58 had delivered vaginally, 12 (20.68%) underwent a caesarean hysterectomy and eight cases (13.7%) delivered by outlet forceps extraction. There were 27.6% fresh still births and 31% early neonatal deaths in the study group (41.4%) were live births. We identified anaemia in 42 cases (72.4%), antepartum haemorrhage in 10 cases (17.2%) as the major risk factor in our series. There were 14 cases (24.1%) each of mismanaged IIIrd stage or injudicious use of oxytocics at peripheral centres. Multiple pregnancy, polyhydramnios, precipitate labour, malpresentations, fibroids and placenta accreta were other risk factors which are outlined in Table 3. Forty nine patients (84.4%) underwent a subtotal hysterectomy, 9 (15.5%) underwent a total hysterectomy and 11 patients (18.9%) underwent a bilateral hypogastric artery ligation additionally.
TABLE 1 :
Correlation with age
No . of cases
< 20 years
> 30 years
TABLE 2 :
Correlation with parity
No . of cases
TABLE 3 :
Identifiable risk factors in this series
No . of cases
Mismanaged IIIrd stage
Injudicious use of oxytocics
TABLE 4 :
Post operative morbidity
Acute renal failure
Post partum psychosis 2 3.41%
The maternal morbidity in the study group included fever (27.5%) acute renal failure (13.7%) and consumptive coagulopathy (18.9%) in terms of medical complications as outlined in Table 4. There were two cases of post partum psychosis. Paralytic ileus was the commonest surgical complication (18.96%). There were two cases of bladder injury and no ureteric injuries in the series of 58 patients.
When medical treatment fails to control postpartum haemorrhage, an emergency hysterectomy may become the necessary evil. In the present study series, aggressive medical management had been tried in all the 58 patients with the use of intravenoxis ergometrine, or oxytocin; and intramuscular administration of analogues of prostaglandin F2. Uterine packing had been done in two cases. However, there is no evidence to prove that uterine packing is of any value. Henson et al, 1983 reported the use of intravenous administration of prostaglandin E2 in persistent primary postpartum haemorrhage.  But since these drugs decrease peripheral resistance, the authors warned that hypovolaemia must be corrected before the drug is given. Hankins et al in 1988 also noted a marked arterial desaturation with analogues of prostaglandin F2a but it was not clear whether this temporary phenomenon was due to the drug or other factors. 
An emergency hysterectomy may be the most effective method for dealing with haemorrhage due to persistent uterine atony or a morbidly adherent placenta. These were the commonest indications cited by Giwa-Osagie et al, Clark et al, Sturdee and Rushton.  ,  ,  Though emergency hysterectomy should be reserved for the management of conditions where conservative measures have failed, its performance should not be delayed until the patient is in extremes.
In the present series, there were two cases of placenta accreta and both required a total hysterectomy. Weckstein also states that total hysterectomy is more frequently performed in cases of placenta accreta, a condition which is becoming increasingly important.  However for other indications of exclusive postpartum haemorrhage, subtotal hysterectomy is the operation of choice, with relatively less blood loss and operative time.  The risk of neoplasia developing in the cervical stump several years later is not relevant in the context of life threatening haemorrhage 72.4% of patients in the present series had been emergency admissions referred from peripheral hospitals and in a moribund post partum hypovolaemic state; hence internal iliac artery ligation could not be attempted as the primary life saving resort without performing hysterectomy. After selective individualization of the case, hypogastric artery ligation is an effective procedure for uterine atony. However, Clark et al, found the procedure to be effective in controlling bleeding in less than half of their 19 patients of uterine atony. They felt, that the complications associated with the procedure were due to a delay in carrying out the definitive treatment (hysterectomy) rather than due to the procedure itself.  Pearse et al reviewed the use of military antishock trousers (MAST) suit. They reported a rapid increase in blood pressure of 65 mmHg, reduced active bleeding and improved perfusion of vital organs. The suit is advantageous for use in the transport of haemorrhaging patients before definitive treatment.  However it may not be easily available in all centres. In conclusion, primary preventive strategies against massive atonic postpartum haemorrhage includes early antenatal registration, prompt identification of high risk factor like anaemia, pregnancy induced hypertension and early referral from primary health centres. At a secondary level, proper management of the 2nd stage, judicious use of oxytocics and careful vigilance of the 3rd stage are vital to prevent excessive bleeding. At the tertiary level, prompt restoration of blood volume after an accurate estimation of the loss is necessary. Quick definitive therapy and performance of surgery by experienced staff would give a favourable outcome. A standardized management protocol implemented smoothly would go a long way in preventing maternal deaths from massive haemorrhage.
We thank Dr. (Mrs.) VR Badhwar, Professor and Head of Department of Obstetrics and Gynaecology and Dr. (Mrs.) AR Fernandez, Dean, LTMG Hospital for permitting us to use hospital data for publication.
1. Combs CA, Murphy EL, Laros RK. Factors associated with postpartum haemorrhage with vaginal birth. Obstet Gynecol 1991; 77 : 69-76.
2. Henson G, Gough JD, Gillmer MDG. Control of persistent primary postpartum haemorrhage due to uterine atony with intravenous prostaglandin E2. Case report. Br J Obstet Gynaecol 1983; 90 : 280-2.
3. Hankins GDV, Berryman GK, Scott RT, Hood D. Maternal arterial desaturation with 15-methyl prostaglandin F2a for uterine atony Obstet Gynecol 1988; 72 : 367-70.
4. Giwa-Osagie OF, Uguru V, Akinla O. Mortality and morbidity of emergency obstetric hysterectomy. J Obstet Gynaecol 1983; 4 : 94-6.
5. Clark SL, Sze-Ya Y, Phelan JP, Bruce SR, Paul RH. Emergency hysterectomy for obstetric haemorrhage Obstet Gynecol 1984; 64 : 376-80.
6. Sturdee DW, Rushton DI. Caesarean and postpartum hysterectomy 1968-1983. Br J Obstet Gynaecol 1986; 93 : 270-4.
7. Weckstein LN, Masserman JSH, Gaeite TJ. Placenta accreta : a problem of increasing clinical significance. Obstet Gynaecol 1987; 69 : 480-2.
8. Lees DH, Singer A. A colour atlas of gynaecological surgery. Surgery of conditions complicating pregnancy. Wolfe Medical, London. 1982; 6 : ...??
9. Clark SL, Phelan JP, Sze-Yay, Bruce SR, Paul RH. Hypogastric artery ligation for obstetric haemorrhage. Obstet-Gynaecol 1985; 64 : 376-80.
10. Pearse CS, Magrina JF, Finley B. Use of MAST suit in obstetrics and gynaecology. Obstet Gynecol Surv 1984; 39 : 416-22.