SAMJ VOLUME 68 23 NOVEMBER 1985 799
The incidence of major abdominal surgery
after septic abortion - an indicator of
complications due to illegal abortion
A. RICHARDS, E. LACHMAN, S. B. PITSOE, J. MOODLEY
Patients and methods
The hospital records of all women admitted with a diagnosis of
That there is a high morbidity associated with illegal 'abortion' were reviewed for the 15-month period 1 June
abortion is well known, but its exact extent is difficult 1983 - 31 August 1984. An abonion was considered to be
to assess, since detailed statistics are not available septic or potentially septic when at least two of the following
and the nature of the morbidity is diffuse. We have four criteria were present: (I) pyrexia > 37,5°C; (il) foul-
assumed that abortal sepsis is a consequence of smelling products of conception; (iil) pelvic peritonitis; and
illegal interference with pregnancy. The incidence of (iv) evidence of lower genital tract trauma (vaginal lacerations,
major abdominal surgery after a septic abortion is cervical tears).
used as an indicator of serious complications. This The number of laparotomies in these cases was recorded
incidence is compared with that following legal abor- and each case reviewed separately. The main indications given
tion. for surgical intervention were: (I) pelvic peritonitis which did
not respond to medical therapy; (il) septic shock resistant to
S AfT Med J 1985; &8: 799-800. active resuscitation; (iil) colpopuneture productive of pus; (iv)
pelvic masses; and (v) uterine perforation and/or uncontrollable
Hysterectomy was performed for: (I) uterine necrosis, chemi-
calor septic; (il) infected uterine perforation; and (iil) uncon-
A number of recent publications have reported a dramatic trollable uterine haemorrhage.
decrease in the morbidity and monality rates associated with
termination of pregnancy, I-3 but these figures are from countries
where abonion is legal and performed in well-equipped
hospitals by experienced medical personnel. In the RSA,
A total of 2450 patients were admitted with the diagnosis of
where the laws regarding abonion are very conservative,
abortion during the 15 months of the study. Of these, 647
criminal abortions with a high incidence of associated morbidity
(26,4%) satisfied the criteria for septic abortion. Fig. 1 demon-
and monality are still common. 4
strates the incidence of uncomplicated and septic abortions in
It is recognized that legal and spontaneous abortions rarely
different age groups. The incidence of septic abortion is not
become infected; a sepsis rate of less than 1% reponed by
statistically different between the age groups, although the
many authors suppons this fact. 3,4 Statistics from King Edward
younger patients tended to have a higher sepsis rate. The
VIII Hospital, Durban, for 1983 reveal that approximately
mean gestational age of the fetuses was 15,2 weeks.
25% of all abonions in patients admitted are septic. Therefore,
Laparotomy. Of the 647 patients with a septic abonion, 42
although a history of interference is seldom obtained it is
underwent laparotomy, giving an operative rate of 6,49%. A
assumed that in the majority of cases abonal sepsis is the
hysterectomy was necessary in 35 patients, 18 of whom were
consequence of criminal interference.
Causes of death after septic abonion are well documented
and include haemorrhage, septic shock and renal failure. How- o
ever, the associated morbidity, panicu1ar1y in terms of the
need for hysterectomy with subsequent loss of fertility, is less
This repon documents the incidence of abdominal surgical
procedures carried out for complications of septic abonion at
King Edward VIII Hospital, which serves as a referral centre
for approximately 5 million blacks in Natal. The majority of
patients come from the Durban metropolitan area, but the
hospital also drains the rural homelands.
Department of Obstetrics and Gynaecology, University of
Natal and King Edward vm Hospital, Durban < 20 years 20-25 ye.,. 26-30 yea" 31·40 yeafs >40 veillS
A. RICHARDS, M.R.C.O.G.
E. LACHMAN, M.D. Fig. 1. Incidence of total abortion, septic abortion and deaths
S. B. PITSOE, F.C.O.G. (S.A.) after septic abortion in the different age groups. The septic
J. MOODLEY, M.R.C.O.G., F.C.O.G. (SA) abortions and total deaths are recorded as a percentage of the
total abortions for each age group.
800 SAMT DEEL 68 23 NOVEMBER 1985
primigravid. Eight patients were under the age of 20 at the
time of hysterectomy.
Mortality. Twelve patients died, despite active management TABLE I. COMPARISON OF ABDOMINAL SURGICAL
- a mortality rate of 1,8%; 7 of these patients had undergone PROCEDURES AFTER LEGAL AND ILLEGAL ABORTIONS
hysterectomy. Their age distribution is shown in Fig. I. Statistical
Age, parity and length of hospital stay. The mean age Surgical significance
of patients undergoing laparotomy was 24,4 years, varying No. of procedure against
between 16 and 42 years; parity ranged from 0 - 6, mean 2,5. Author patients rate (%) present series
The mean hospital stay was 14,6 days, (range 2 - 57 days) Grimes et al.' 136056 0,027 p= 0,001*
compared with a mean stay for non-septic abortion patients of King eta/} 11885 0,067 p= 0,001*
I or 2 days. Hodari et al. 3 2500 0,12 p= 0,001*
Indications for surgery. Sepsis was the main indication Wulff and Freiman 7 16410 0,03 p= 0,001*
for surgery (32 cases). In a further 6 cases the indication was
Present series 647 6,49
uterine trauma and/or uncontrollable uterine bleeding. In the
... For each series reported. the difference in incidence was found to be statistically
remaining 4 cases the indication was not clearly stated.
significant in respect of our series for septic abortion.
Abortion is the most· common indication for gynaecological
admissions to King Edward VIII Hospital, and although I), thus demonstrating the relative safety of legal abortion.
criminal abortion is common a positive history is seldom Further evidence is given in the Report on Confidential Enquiries
obtained. into Maternal Deaths in England and Wales 1976-1978,6 which
From the few histories available it appears that abortion clearly demonstrates a dramatic decrease in fatalities due to
methods commonly used include intra-uterine injection via a criminal abottion since the introduction in Britain of the
catheter of chemicals such as cWoroxylenol, washing blue or Abortion Act of 1967.
soapy solutions. Patients have also described the use of herbal Our data confirm a high incidence of serious complications
enemas, knitting needles, oral quinine, castor oil and laxatives. after illegal abortion; treatment involves the expenditure of
Septic abortion requires active management in terms of great sums of public money. More extensive morbidity in
resuscitation, intravenous antibiotics and uterine evacuation. terms of chronic pelvic inflammatory disease, tubal damage,
It is suggested that a colpopuncture be performed at the time infertility and psychological trauma is difficult to assess,' but
of uterine evacuation in all cases of septic abortion. A positive it also takes a significant toll in terms of suffering and human
colpopuncture for pus, blood or abnormal fluid would suggest misery.
intra-abdominal sepsis, uterine ttauma or chemical necrosis, The need for family planning and sex education in the
and a laparotomy would be indicated. Further indications for community must be emphasized; a reappraisal of the law on
laparotomy are peritonitis, septic shock and uncontrollable abortion in the RSA is suggested.
uterine bleeding. Early hysterectomy can prevent later deteri-
oration due to endotoxic shock and multi-organ failure. In
certain cases with obvious uterine necrosis, as indicated by REFERENCES
gangrene of the cervix, initial uterine evacuation is contra-
1. Grimes DA, Flock HL, Schulu KF, Cates W. Hysterectomy as rrearment
indicated. for complication of legal abortion. Obsrec Gynecol 1984; 63: 457-462.
Our study confirms a high incidence of abdominal surgical 2. King TH, Atienza MF, Burkman RT. The incidence of abdominal surgical
procedures in a population undergoing abortion. Am] Obsrer Gynecol 1980;
procedures after abortal sepsis. Thirty-five hysterectomies were 137: 530-533.
necessary; these proved to be life-saving in only 28 cases. The 3. Hodari AA, Peralta J, Quiroga PJ et al. Dilatation and curertage for second
majority of hysterectomies were done on young patients with trim,ster abortions. Am] Obsccr Gynecol 1977; 127: 850-854.
4. Lachman E, Pitsoe SR, Gaffin SL. Anti-lipopolysaccharide immunotherapy
no children. in the management of septic shock of obsterric and gynaecological origin.
In the recent study of legal abortions by Grimes et al. 1 in Lancer 1984; i: 981-983.
5. Wulff GJL, Freiman SM. Elective abortion, complications seen in a free-
the USA, hysterectomy was used as an indicator of serious standing clinic. Obsccc GynecoI1977; 49: 351-357. .
complications. These authors report a decline in the rate of 6. Thomkinson J, TumbuU A, Robson G et al. Reporr 011 Confidenrial Enquiries
inco Macemal Deachs in England and Wales 1976 -1978 (Department of
hysterectomy after legal abortion from 0,45% in 1970 to 0,014% Health and Social Security). London: Her Majesty's Stationety Office, 1982.
in 1978. Other authors 2,J,5 have confirmed these fmdings (Table 7. Larsen JV. Induced abortion. S Afr Med] 1978; 53: 853-857.