COMPARISON OF PRE-OPERATIVE PERITONEAL WASH AND DRAINAGE VERSUS URGENT LAPAROTOMY STRATEGY IN PATIENTS WITH LATE STAGE PERFORATION PERITONITIS . A RANDOMIZED CONTROLLED TRIAL

Peritonitis is inflammation of peritoneum which is most commonly due to generalized or localized infection. Secondary peritonitis is the commonest form and a large percent of them is due to perforation or impending gastrointestinal perforation. The prognosis depends on multiple factors by which Mannheim Peritonitis Index (MPI) appears to be more practical. This study aimed to assess the value of pre-operative aspiration of peritoneal fluid followed by peritoneal wash and drainage before proceeding to definitive surgery in patients with MPI score>20. This prospective study was conducted in Al-Hussein Teaching Hospital in Al Nasserya city and Basrah Teaching Hospital in Basrah city from October 2003 to July 2014. Each patient admitted to the emergency department in these hospitals with the provisional diagnosis of perforation peritonitis was evaluated with MPI score. If the score was less than 20; the patient managed with resuscitation and broad spectrum antibiotics for 2-3 hours then by definitive surgery while those with MPI more than 20 were randomly divided into two groups; the first were managed with 2-3 hours resuscitation with intra-venous fluid resuscitation and antibiotics followed by urgent surgical exploration (USE). The second group were managed with percutaneous peritoneal drainage (PPD) with aspiration of the fluid and then irrigation of the peritoneal cavity with isotonic saline and followed by a drainage with aid of the gravity through another catheter located in the right ileac fossa. Sixty two patients included in this study who fulfilled the criteria of perforation peritonitis and MPI score more than 20. Around half of them the score was between 26-30. Perforated peptic ulcer is the commonest etiology. The most common cause of their high MPI score is the late presentation and the evidence of organ failure. Renal failure is the most prevalent organ failed in both groups. The overall mortality is decreased in (PPD) group. Those with urgent surgical exploration(USE) showed no improvements in the pre-operative vital signs, prolonged operation time and a higher mortality and more severe post-operative complication than PPD group. In conclusion, the pre-operative percutaneous peritoneal aspiration of the fluid followed by peritoneal irrigation and drainage in patients with advanced stage perforation peritonitis is associated with a significant improvement in the pre-operative pulse rate and blood pressure, decrease in the operation time, a decrease in the overall mortality and deep seated wound infection and dehiscence but it is associated with a higher mortality in the first post-operative day. Introduction P eritonitis is inflammation of peritoneum which is most commonly due to generalized or localized infection. Secondary peritonitis is the commonest form. A large percentage of secondary peritonitis is due to perforation which if not treated surgically causes considerable mortality. Mortality of perforation peritonitis was as high as 90% in the early twentieth century and is still high despite advances in antibiotics, surgical technique, radiographic imaging, and Bas J Surg, December, 24, 2018 67 Comparison of pre-operative peritoneal wash and drainage versus urgent laparotomy Salah Kadhim Muslim resuscitation therapy. The contamination with the peritoneal cavity can lead to a cascade of infection , sepsis , multi system-organ failure (MSOF) and death if not treated in a timely manner. Many scoring systems was used to assess the severity of perforation peritonitis like Acute Physiology And Chronic Health Evaluation score(APACHEII), Simplified Acute Physiology Score (SAPS). In 1983; Wacha and Linder developed a scoring system consist of twenty risk factors on which eight of them proved to be of prognostic relevance. This system called Mannheim Peritonitis Index (MPI) (Table I). It appears to be more practical than other scoring systems. Mortality increases with increasing range of the score, adopting three cut-off ranges from less than 20, from 20 to 30 and more than 30 growing the mortality from 0% to 28% up to 81% respectively. Table I: The Mannheim peritonitis index Risk factor score


Introduction
P eritonitis is inflammation of peritoneum which is most commonly due to generalized or localized infection.Secondary peritonitis is the commonest form 1,2 .A large percentage of secondary peritonitis is due to perforation which if not treated surgically causes considerable mortality.Mortality of perforation peritonitis was as high as 90% in the early twentieth century and is still high despite advances in antibiotics, surgical technique, radiographic imaging, and resuscitation therapy.The contamination with the peritoneal cavity can lead to a cascade of infection , sepsis , multi system-organ failure (MSOF) and death if not treated in a timely manner 3,4 .Many scoring systems was used to assess the severity of perforation peritonitis like Acute Physiology And Chronic Health Evaluation score(APACHEII), Simplified Acute Physiology Score (SAPS).In 1983; Wacha and Linder developed a scoring system consist of twenty risk factors on which eight of them proved to be of prognostic relevance.This system called Mannheim Peritonitis Index (MPI) (Table I) 4,5 .It appears to be more practical than other scoring systems.Mortality increases with increasing range of the score, adopting three cut-off ranges from less than 20, from 20 to 30 and more than 30 growing the mortality from 0% to 28% up to 81% respectively 1,4,5 .In order to decrease the mortality and morbidity of a patient with perforation peritonitis, adequate fluid resuscitation should be started, initiation of antibiotics and then source control in the form of emergency surgery.The principles of source control dictate direct control of the site of perforation, evacuation of intra-peritoneal contamination, drainage of abscesses, debridement of necrotic tissue and foreign matter, and reestablishment of functional anatomy 6 .If there is delay in the initial management; many of patients developed systemic inflammatory response syndrome(SIRS), sepsis and septic shock with accompanying hemodynamic compromise, hypothermia, acidosis, and a coagulopathy.The challenge is how to manage such patients without causing further physiologic compromise.So, source control should be achieved with the least invasive maneuver [7][8][9] .Sepsis source control through percutaneous peritoneal drainage is not new.Atakent YS et al and Lessin MS et al and others described a definitive treatment for acute intestinal perforation in extremely very low birth neonate by percutaneous peritoneal drainage [10][11][12] .Ramaswamy and Niall et al discuss the same principle in the treatment of bladder perforation during transurethral resection of prostate 13 .S.Peterson-Brown and H.A.F Dudly stated that "in a circumstances of severe abdominal distension, peritoneal aspirate should be considered by peritoneal dialysis catheter in order to permit the escape of gas and exudate, thereby providing an initial guide to the diagnosis in addition to relieving the cardiopulmonary effect of distension".They also stated that "if the patient's condition still prevents operation, treatment may began by running a liter of normal saline with or without antibiotics over a period of an hour and then recovering it by gravity drainage" 14 .The aim of this study is to assess whether pre-operative drainage of purulent peritoneal fluid followed by peritoneal wash and drainage can decrease the mortality and the morbidity of patients with perforation peritonitis who had MPI score more than 20.

Patients and methods
This prospective study was conducted in Al-Hussein Teaching Hospital in Al Nasserya city and Basrah General Hospital in Basrah city from October 2003 to July 2014.Each patient admitted in the emergency department of these hospitals with the provisional diagnosis of perforation peritonitis was re-evaluated by the resident surgeon on-call and then by the same specialist surgeon who scored each patient according to MPI index.Patients with peritonitis secondary to esophageal perforation and reproductive tract perforation and those who had acute abdominal pain of uncertain etiology were excluded from the study.Each patient with MPI index less than 20 were managed by rapid fluid resuscitation, broad spectrum antibiotics for 2-3 hours and then definitive surgery while those with index more than 20 were randomly divided into two groups of management; those presented on odd numbered days were managed with resuscitation followed by urgent surgical exploration (USE) and those who presented on even days were managed with percutaneous peritoneal drainage (PPD) which is performed by the following steps: 1-Informed verbal consent taken from the patient's family with full details about the procedure.

2-
The patient was kept in supine position with painting the skin around the umbilicus followed by draping then, a 5-10 milliliters of 1% xylocaine as a local anesthesia was infiltered subcutaneously then a one centimeter incision in the skin just above the umbilicus followed by insertion of a 16 or 18G peritoneal dialysis catheter directed toward the pelvis, aspiration of the fluid(if any) found in the peritoneal cavity in 10 ml syringe.The fluid drained was inspected for the color and sent for Gram stain and culture.Another tube drain (14 or 16 G)was inserted through a separate incision in right ileac fossa under local anesthesia as a pelvic drain.3-A 1000 ml of warm isotonic saline was infused into the peritoneal cavity through peritoneal dialysis catheter with occlusion of pelvic drain for 30 minutes.Then release the occlusion and drain the wash through the pelvic tube drain with changing the patients' position to more head up position.This step was repeated each 3-4 hours for 24 hours.During that, patient's response was observed with a follow up chart in the ward or in ICU focusing on the conscious level, vital signs, urine output and oxygen saturation by pulse oxymetry.After that, the abdomen was then explored for definite procedure by the same surgeon.Data were recorded including patients age, sex, pre-operative vital signs, the color, and the culture of peritoneal aspirate, the definitive diagnosis after surgical exploration, the duration of definitive surgical procedure and postoperative morbidity and mortality.Chisquare (χ2) test was used to compare variables and tests were considered significant when P-Value ≤0.05.

Results
Sixty two patient presented to the emergency department with a definitive diagnosis of perforation peritonitis and MPI index more than 20.They were 38 males and 24 females.Their age ranged between 14-90 years with a mean age of 39.7 years.They were randomly classified into two groups; percutaneous peritoneal drainage (PPD N:32) and urgent surgical exploration (USE N:30) as shown in table II.

Table II: gender distribution of the patients in the two groups.
Late patients presentation to the surgical emergency department or to the surgeon, ranging between 4-17 days as shown in figure 1.In both groups, the patients were further sub-classified into three groups according to MPI score (figure 2).It shows that around half of the patients presented with MPI between (26-30).Each patient was evaluated for evidence of organ failure at times of presentation which was discovered in 58 (93.5%)patients.Table III, shows the site of organ failure associated with both groups.Renal failure is the most prevalent organ failed in both groups (50%).There was no patient who had a features of bone marrow failure or hepatic failure.Pre-operatively, a sample from the peritoneal fluid could be aspirated in 49( 79%) patients which was then inspected for the color (figure 3).In all patients, and on surgical exploration, a sample from the peritoneal fluid was aspirated and then sent for culture as shown in figure 4. It is clear that the most common bacteria cultured from the aspirate was E.coli with or without other enteric bacteria.Negative culture was detected in 26 (42%) patients.The changes in the vital signs were evaluated on admission and then checked just before surgical interference in both groups as described in Table IV.There were statistical significant decline in the pulse rate and increase in the blood pressure in PPD group only.
In both groups, an intravenous anti-biotic cover was started on admission triple antibiotics (ampicillin QDS, gentamycin TDS, metronidazole TDS) or third generation cephalosporin BID with metronidazole.The patients were finally explored through midline laparotomy for definitive diagnosis and management.Figure 5 shows the underlying causes of perforated peritonitis in all patients.The average time of the whole surgical procedure was calculated in both groups.In PPD group it ranged between 50-130 minutes with a mean time 61.2 minutes.In USE group it ranged between 50-120 minutes with a mean time 83 minutes which shows a statistical significant reduction in operative time in PPD group as shown in figure 6. Overall mortality in both groups was 35.4% (22 patients):40% (12) in USE group and 31.2%(10) in PPD group.This differences is statistically significant.Figure7 shows the percent mortality in association with time of death.Most of deaths (70%) in PPD group were occurred in the first 24 hours.In USE group; although the highest mortality was in the first 24 hours postoperatively, it is clear that the patients still had a significant mortality after that with a maximal incidence in the second post operative day.All patients in both groups who died in the first 24 hours after surgery show no recovery from anesthesia and they continue on mechanical ventilation after surgery till development of a cardiac arrest and death.In PPD group; those who survive the first 24 hours showed a significant decrease in mortality which was mostly due to a complication of organ failure; while those in USE group showed persistent risk of death from a complication of organ failure and sepsis from residual septic foci as shown in table V.It is clear that 40% of patients in USE had postoperative complications compared to 37.5% in PPD group.In addition, the latter group showed a higher rate of superficial surgical site infection and a lower rate of deep seated wound infection and dehiscence than USE group.On comparing the percent mortality in association with MPI score; it is found that there is no statistical significant differences between the two groups as in figure 8.

Discussion
In the first quarter of the twentieth century; Martin Kirschner first described the standard surgical approach for treatment of perforation peritonitis 15 .It consist of source control, reduction in bacterial contamination and prevention of its recurrence 7 .Other measures are of little use if the operation does not successfully abort the infective source and quantitatively reduce the inoculation of micro-organisms and adjuvant of infection so that they can be effectively handled by the patient's defenses, supported by antibiotic therapy 7,[15][16][17][18] In this study, patients with advanced peritonitis were selected depending on their high MPI index.The cause of which is mainly due to the delay in presentation and evidence of organ failure.There are multiple factors explain such delay; either the patients ignored the earlier symptoms, or had taken herbal or medicinal care by local health facility, or those patients lived in places far from centers with surgical facilities, so had to travel long distances to reach a referral center.Such delay leads to widespread dissemination of the primary insult making the control of pathology difficult and resulting in poor intraoperative outcome.It could be regarded as the most important factor predicting in the prognosis of peritonitis especially perforation peritonitis 6,12 .
In this study, renal system is the most common organ failed followed by cardiopulmonary system.These findings are different from other reports like that reported by Ali Yaghoobi etal and Betra who found that the most common organ failed is cardiopulmonary followed by the hepatic failure 4,5 .Exact explanation for such differences is unknown.Gastro-intestinal failure lacked clear definition, its incidence was low and its occurrence was rarely associated with poor outcome.This study reported also that failing central nervous system is associated with other organ failure, high MPI score and a significant mortality 4 .Although E.coli and other Gram -ve enterococci are considered the most common intra-abdominal pathogens, the data presented here indicate that these are the GIT flora and much more causing peritonitis when spilled to the peritoneal cavity converting into pathogenic one 19 .Also because these micro-organism are more prone to determine septicemia for their high affinity to reach the blood circulation .Anaerobic and opportunistic bacteria usually leads to circumscribed peritonitis 1 .Negative cultures were found in around one third of our patients; a finding which is similar to others 4,5 and can be explained either by the chemical rather than bacterial nature of peritonitis that occurs specially with upper GIT perforation before proceeding to a secondary bacterial peritonitis or due to the effect of pre-operative use of antibiotics 20 .During pre-operative resuscitation in both groups; a statistically significant change in both pulse rate and in systolic blood pressure were observed in PPD group while no significant improvement was reported in the temperature and respiratory rate.These findings although were not reported by the others; but clearly indicates that the 24 hours of resuscitation elapsed between the diagnosis of perforation peritonitis and the definitive source control was enough to make an apparent improvement in the pulse rate and systolic blood pressure but it was an inadequate time for the thermo-regulatory and acid-base balance which needs more time to show a significant changes.In addition; the cause of tachypnea which is mainly due to lactic acidosis and the source of hyperthermia which is the pyrogens and the inflammatory mediators released from the inflammatory cells which is not eliminated during that time 19 .On surgical exploration, the commonest underlying etiology is perforated peptic ulcer followed by perforated acute appendicitis.This finding is consistent with other researches in the nearby countries 21 while different from other researches done in Europe, USA and Japan in which acute perforated appendicitis is the most common cause followed by colonic perforation mostly due acute diverticulitis 4,5,19,20,22 .There is a significant drops in the incidence of proximal bowel perforation in the developed nations due to the better availability of proton pump and adoption of therapies against Helicobacter pylori.Distal bowel perforations especially colonic perforations are the leading cause of perforation peritonitis in the western world 2,23,24 .Regarding the time needed to complete the surgical procedure , there is a statistically significant decrease in the time in PPD group in which there is less time of initial suction of peritoneal fluid and less peritoneal wash after completion the source control.This is because that some of these actions were already started pre-operatively.There is a wide range of in-hospital mortality of patients with perforation peritonitis(3%-58%) 4,18,22,23 , but the mortality rate among patients with high severity scores is ranging between (34-58%) [23][24][25][26][27][28] .This study results are consistent with this range in both groups but there is a statistical significant decrease in overall mortality in PPD group. in the first 24 hours after surgical exploration, the mortality rate of PPD group is more than USE group in which the patients did not recover from anesthesia.This finding may be explained by the prolonged effect of the myocardial depressant factors like tumor necrotic factor-alfa (TNF-alpha), platelet activating factor (PAF), gamma interferon (IFN-gamma) and arachidonic acid metabolites which can continue for days and further depressing cardiac function even with adequate intravenous fluid therapy keeping in mind that before anesthesia; sufficient preoperative systemic blood pressure is not indicative of adequate volume status, and therefore, heart rate, urine output, and mental status should all be considered when evaluating the volume 7 .Additionally, there is a risk although theoretical that the increase intra-abdominal pressure created by both pre-operative peritoneal lavage and the positive pulmonary ventilation during anesthesia will increase the opening of special holes found in the peritoneal surface of the diaphragm called lymphatic lacunae through which bacteria and toxins can pass directly to the blood stream through the thoracic duct.This complication can be minimized by frequent suctioning of peritoneal fluid with the least volume of peritoneal lavage.While peritoneal lavage for peritonitis is universally recommended, remarkably little studies has been done on its specifics.Too many articles on peritoneal lavage in peritonitis are available but only one small clinical control trial by Schein in 1990, comparing no lavage with intra-operative lavage with and without antibiotics, showed no differences in survival.As a result, Schein recommends only "swabbing or mopping peritoneal surfaces with moist laparotomy packs" [28][29][30] .Despite this, intra-operative lavage remains standard therapy.It is recommended that all fluid be aspirated at the closure of the abdomen as there is evidence that the ongoing presence of fluid decreases macrophage effectiveness.Studies looking at post-operative continuous lavage showed limited success with increasing risk of fistulization 7,29 .The latter surgical complication was also associated with radical debridement of fibrinous exudates.One trial showed it to be more dangerous than lavage alone and it has been abandoned.Addition of antibiotics to lavage solution also has not been shown valuable, neither has postoperative continuous lavage or use of drains 18 .Post-operatively, both groups showed no statistical significant differences in surgical complications in general, but USE group showed a statistically significant higher incidence of deep seated wound infection and intra-abdominal abscess and hence burst abdomen than PPD group.This finding is explained by the prolonged reduction of bacterial contamination in PPD group which was started before definitive surgical repair during which the time was spent for only opening the contaminated spaces, aspirating the remaining purulent fluid and removing food, feces and foreign debris while in USE group all these actions should be done in limited time in addition to peritoneal lavage.Conclusions, the pre-operative percutaneous peritoneal aspiration of the fluid and peritoneal irrigation and drainage in patients with advanced stage of perforation peritonitis documented by their high MPI score for 24 hours before definitive surgical source control is associated with a significant improvement in the pre-operative pulse rate and blood pressure , decrease in the operation time, a decrease in the overall mortality and deep seated wound infection and dehiscence but it is associated with a higher mortality in the first postoperative day indicating that the 24 hours of preoperative wash and drainage is not enough to eliminate all the factors that are responsible for increasing mortality.

Figure: 1 :
Figure:1: Time of first presentation to the surgical emergency department in days in both groups.

Figure 2 :
Figure 2: The MPI score in both groups.

Figure 3 :
Figure 3: The color of the aspirated fluid.

Figure 4 :
Figure 4: The result of aspirated peritoneal fluid culture .

Figure 5 :
Figure 5: the definitive diagnosis in both groups .

Figure 6 :
Figure 6: The differences in the meantime of surgery between the two groups.

Figure 7 :
Figure 7: The percent mortality in the post -operative days.

Figure 8 :
Figure 8: Percentage of mortality in MPI score in both groups.

Table III : Organ failure in both groups.
Multiple failure seen in six patients of PPD group and in three patient of USE group.All patients with CNS failure were those with MPI score more than 30 and associated with other organ failure.

Table IV : The changes in mean vital signs on admission and just before operation in both groups.
** statistically significant at p ≥0.05 º statistically not significant.

Table V : The cause of death in both groups in the post-operative period
After recovery from anesthesia, all patients were kept on a closed and a regular observation chart for any evidence of surgical complications which is seen 12 patients in each group.TableVIshows the surgical complications observed in both groups.