Fatal perforative peritonitis - (study of 38 cases)Sulabha V Punekar, CV Patel, GB Parulkar
Department of Surgery, Seth G. S. Medical College and K. E. M. Hospital, Parel, Bombay-400 012, India
Correspondence Address: Source of Support: None, Conflict of Interest: None PMID: 615256
Source of Support: None, Conflict of Interest: None
The etiology, clinical features and treatment of thirty-eight cases of fatal perforative peritonitis were studied. The commonest cause of fatal peritonitis in this series was found to be ileal perforation. Early diagnosis, intensive post-operative management including fluid and electrolyte balance, and employment of potent antibiotics like gentamycin and kanamycin would definitely help in reducing mortality of peritonitis.
The mortality of perforative peritonitis presenting itself as an acute surgical emergency remains about 30% in India. There are several reports published of peritonitis as general or individual perforation. Very few studies have been done on only fatal cases of peritonitis. The intention of the present study was to investigate the aetiology of fatal peritonitis so that a better treatment could be planned.
The total number of surgical admission; into The K.E.M. Hospital, Bombay during 1973-1975 was 22833. Out of these, 297 were admitted as perforative peritonitis Thirty-eight cases of peritonitis who died before or after operation were selected for the present study. Their clinical features results of laboratory investigations, operative findings and autopsy data were critically analysed retrospectively.
1. Age: Patients ranged in age group of 10 days to 75 years. Maximum number of patients (17 out of 38) were in the Group of 20-40 years.
2. Sex: Twenty-seven were males and eleven females, the male to female ratio being 2.5: 1.
3. Duration of Symptoms: Varied from 1 day to 6 weeks, but the majority of patients had symptoms for 1-7 days.
4. Symptoms: The commonest symptoms were as follows:
(a) Abdominal Pain No.of.Patients -38, Percentage -100
(b)Fever,vomiting abdominal distension No.of Patients -19,Percentage -50
(a) Abdominal tenderness - 38 -100
Pulse 100-120/ mt. - 31 - 90
Pulse 120-140/mt. - 20 - 58
(c) B.P. (Below 80 mm, Hg. Syst.) - 10 - 26
(d) Signs of peritonitis - 19 - 50
(a) X-ray of Abdo- No. of Positive men (for Gas Patients in under Diaphragm) - 38 - 14(36%)
(b) Peritoneal tap - 12 - 9(75%)
(c) The routine blood counts were done in 26 cases but not in all cases because of the emergency nature of the illness. Haemoglobin estimation was done in 26 cases and the range was 6 to 13 Gms.% . Serum electrolyte readings showed hypokalaemia (K-2-3 mEq/L) hyponatraemia (Na-120-130 mEq/L) and hypochloraemia (Cl 80-90 mEq/L) in 30% of cases, most of which were post-operative readings.
(d) Pus from peritoneal cavity collected at the time of diagnostic tap, during operation or autopsy was sent for smear and culture. In the majority of cases the organisms detected were E. Cole, sensitive to gentamycin, ampicillin and kanamycin. In 8 cases Klebsiella and in 2 cases Pseudomonas and Streptococci were grown.
7. Management: Pre-operative supportive therapy including I.V. fluids, antibiotics and blood transfusion was given in all cases. Penicillin, streptomycin, chloramphenicol, tetracyclines were the common antibiotics used alone or in combination. The first three drugs in combination were used in 25 patients while Gentamycin was used in only four patients.
Further treatment was as follows:
Type of Treatment - No. of Patients
1. Exploratory laparotomy and closure of perforation - 26
2. Drainage of peritoneum under local anaesthesia - 4
3. Patients died before operation could be undertaken - 8
8. Duration of Hospitalization: The average survival of patients after admission was 7.5 days (range-3 hours to 33 days).
9. Autopsy Findings: Autopsies were performed in 32 cases. The diagnosis and cause of death were obvious in 6 operated cases and therefore autopsy was not performed in these patients. Various sites of perforations are given in Table below:
The highest age incidence (44%) in the 2nd and 3rd decade found in this study is more or less similar to that reported in other series. The male: female ratio is somewhat different than in other series  but no definite conclusion can be drawn as this series is very small. It is difficult to compare the present series with any one report because it includes only fatal cases of peritonitis. However, individual type of perforation is studied and compared wherever possible.
In other series duodenal perforation was the commonest cause of peritonitis ,whereas in this series it was the second commonest cause (23%). The previous history of ulcer was not present in 66% of cases in this series as compared to 59.5%, in Budharaja et al,  series 60% by Udwadia, et al,  and 75% by James and Mark.  The treatment of simple closure of perforation and drainage of peritoneal cavity was done in 8 out of 9 duodenal perforations it this series. This treatment is preferred to definitive surgery in bad generalised peritonitis cases in the opinion of James and Mark.  Only one patient was not fit for general anaesthesia where peritonea] cavity was drained under local anaesthesia. Ninety per cent of these patients had come to hospital 6-7 days after the onset of acute symptoms. Seven out of nine patients were about 55 years of age. These two causes account for the deaths in duodenal ulcer perforation in this series.
Ileal perforation was the commonest cause in this series (42%) . In about 62% of these, a history of previous illness was available for 7 to 10 days. The ileal perforation thereby seems to occur at the end of first or beginning of second week, an observation similar to that by other series. ,,, During these first 7-10 days, patients were treated at home and no definite history regarding the names of drugs was available in most of these cases. The use of corticosteroids cannot therefore be ruled out. The Widal Test was positive in only 20% of cases while other authors have reported higher percentage. , Though the site of perforation was similar to that in other series, , the incidence of multiple perforation was much smaller (2.5%) in this study as compared to 5% by Gandhi et al  Twelve out of 16 patients were operated under general anaesthesia. There has been controversy between the conservative treatment as advocated by Huckstep  and surgical closure as advised by Franklin.  The surgical closure and peritoneal lavage appears to be more rational because it prevents the continuous leak from the site of perforation. On the other hand, patient with ileal perforation is in severe fluid and electrolyte imbalance and does not usually stand the strain of prolonged anaesthesia for exploratory laparotomy and closure. Probably an elaborate at- tempt to correct electrolytes with local drainage of peritoneum before surgery would improve patient's general condition to enable him to stand the strain of anaesthesia and operation.
The histopathological studies showed non-specific type of ulcer a little more common (43% ) than typhoid ulcer perforation as compared to other series (23%).  The peritonitis sets in very fast once the perforation occurs in ileum and the diagnosis is usually delayed because of previous illness of patients like diarrhoea, abdominal distension and fever. This is another factor increasing the mortality in ileal perforation.
The incidence of colonic perforation as found on autopsy studies here was 18.5%, as compared to 30.4% by Kean et al,  and 10.5% by Clerk.  The age group 23 to 75 years was similar to that in other series , Amoebic colonic perforation was much commoner cause of perforation yet the clinical picture in these cases appeard to be deceptive, as in four out of seven cases accurate diagnosis was delayed till autopsy. History of diarrhoea was present in 56%, as compared to 45.7%  in other series. The site of colonic perforation was caecum and ascending colon in 66% and rectosigmoid in 33%, similar to those reported by Chen et al .  Multiple perforations were found in 2 patients (8%) as compared to 75% reported by Chen et al.  The highest mortality in colonic perforation may be explained by (1) variability of clinical picture, (2) its frequency in older age groups, (3) multiplicity of perforations, (4) gram negative organisms invading peritoneum. This mortality could be reduced by-(1) Abandoning the conservative management at earlier stage in a doubtful case, (2) drainage of peritoneum with colostomy, (3) diversion of faecal matter, (4) use of higher antibiotics along with antiamoebic drugs.
This series included only one case of amoebic liver abscess rupture. Repeated aspiration, local drainage, drugs like metronidazole and chloroquin have probably caused reduction in mortality in recent years.
Appendicular perforation was less common (3%) in this series as compared to higher figures in other series like Budha raja et al  and Chen et al. 
The culture studies of pus collected from peritoneal cavity showed E. coli growth sensitive of gentamycin, kanamycin, and ampicillin, similar to those reported by Budharaja et al.  The majority of patients were treated with a combination of penicillin, streptomycin and chloramphenicol. Only 4 patients were given gentamycin and ampicillin after initial treatment with 3 antibiotics mentioned earlier, because the cost of higher antibiotics was prohibitive. The use of these broad spectrum antibiotics like gentamycin and kanamycin at art earlier stage in frank cases of generalised peritonitis would have probably helped to reduce mortality. Early diagnosis, intensive post-operative management as regards fluid electrolyte balance and employment of higher antibiotics would help reduce mortality of perforative peritonitis.
Thanks are due to the Dean, K.E.M. Hospital, Bombay 12, for permission for allowing us to use hospital data and to publish this paper.