It is almost a century since Moynihan famously noted that, “every surgical operation is an experiment in bacteriology.”1 While the basic components of that experiment (the host, the bacterial flora and the factors which alter the balance between bacterial capacity for invasion and host resistance) have not changed, the outcome of surgical procedure, even when complicated by infection, has improved markedly. The last century has seen overwhelming advances in reducing the incidence of surgical (and other) infections, as well as an ability to treat them more effectively.
These have resulted from (inter alia), better nutrition and overall public health, a better understanding of factors which contribute to infection such as temperature maintenance, oxygenation, blood glucose control, and aseptic technique, and of course the development of powerful antibiotics for both prophylaxis and treatment of infection complicating surgical care.
However, despite these measures, infection remains a common and life-threatening problem and the sepsis, which results from this infection, continues to be a significant cause of avoidable mortality, morbidity and health expenditure. Some of the most devastating adverse consequences of infection result not from the direct pathogenic effects of the invading bacteria but from the immunological consequences of the host response. While a coordinated pro- and anti-inflammatory response is essential for localisation, bacterial killing, and resolution, infection can trigger an overwhelming host inflammatory response, resulting in shock, multi-organ dysfunction, and death.
The definitions of sepsis, septic shock, and organ dysfunction were based on an international consensus conference,2 which focused on the then-prevalent view that sepsis developed as part of a host systemic inflammatory response syndrome (SIRS), triggered by an infectious insult, noting that sepsis could arise in response to multiple infectious causes and that ‘septicaemia’ was neither a necessary condition nor a helpful term. It was proposed that sepsis complicated by organ dysfunction was termed severe sepsis, which could progress to septic shock, defined as “sepsis-induced hypotension”, persisting despite adequate fluid resuscitation or by hyperlactataemia.
In 2001, a second consensus panel endorsed most of these concepts, with the caveat that signs of a systemic inflammatory response, notably tachycardia or an elevated white-cell count, also occur in many non-infectious conditions and therefore are not helpful in distinguishing sepsis from other conditions.3
In addition, severe sepsis and sepsis were sometimes used interchangeably to describe the syndrome of infection complicated by acute organ dysfunction. Attempts to take account of the fact that critical illness might arise as a consequence of infection, without the requirement for the patient necessarily to exhibit the fever, tachypnoea, tachycardia and leukocytosis required of SIRS led to a third International consensus for sepsis and septic shock, at which sepsis was defined as life-threatening organ dysfunction caused by a dysregulated host response to infection.4
This new definition also recommended using qSOFA (quick Sequential (Sepsis-related) Organ failure Assessment): respiratory rate of 22/min or greater, altered mentation, or systolic blood pressure of 100 mmHg or less, for rapid bedside assessment with higher predictive validity of sepsis than traditional SIRS criteria. Early sepsis-related organ dysfunction was defined as an acute change in total SOFA score ≥ 2 points, because of infection.
The National Early Warning score (NEWS) was first produced in 2012 and updated (NEWS2) in December 2017.5 NHS England and NHS Improvement have endorsed NEWS2, which is based on a simple aggregate scoring system in which a score is allocated to physiological parameters routinely measured in clinical practice – respiratory rate, oxygen saturation, systolic blood pressure, pulse rate, level of consciousness or new confusion and temperature.6 Three of these parameters are same as those of qSOFA and with added parameters from NEWS2, the ability to identify patients at risk of sepsis will be remarkably enhanced.
A raised NEWS of 5 or more should trigger immediate escalation of treatment and patients affected treated for sepsis until proven otherwise, as this failure to do so has been shown to be associated with a >threefold increased risk of transfer to the intensive care unit or death.7
Burden of surgical sepsis
NHS England have estimated that approximately 120,000 patients develop sepsis each year and more than 37,000 people die as a consequence. Sepsis is the second most common cause of death after cardiovascular disease.6 While these data relate to sepsis resulting from all infection, the most common cause of sepsis in surgical patients is intra-abdominal infection, which accounts for approximately two-thirds of all cases.8,9 While sepsis might arise from intra-abdominal (or retroperitoneal) infection resulting from upper gastrointestinal, hepato-pancreaticobiliary and small intestinal disease, colonic (including appendicular) perforation predominates.10
It is sobering to note that Moynihan’s 1920 observations still remain valid, and a significant proportion of hospital-acquired sepsis arises as a direct consequence of complications of abdominal surgery. The National Confidential Enquiry into Patient Outcome and Death (NCEPOD) review Just Say Sepsis11 noted that more than 60% of patients with hospital-acquired sepsis developed their infection as a result of an invasive procedure.12
Despite advances in management, the development of septic shock in a patient with infection still has considerable negative prognostic implications. Septic shock is associated with an overall hospital mortality of 39% in patients admitted as a surgical emergency and 30% mortality in those admitted for elective surgery.9 It is therefore imperative to identify sepsis promptly, commence resuscitation and antimicrobial therapy, and achieve rapid source control.
Current clinical guidelines on the management of sepsis, such as those from NHS England,2 Surviving Sepsis Campaign,13 the English National Institute for Health and Care Excellence14 and the Irish National Clinical Effectiveness Committee,15 all emphasise that sepsis is a medical emergency with a limited window of opportunity for effective intervention. The new sepsis Hour-1 Bundle spells out essential steps that should be undertaken in the first hour of managing a patient with suspected sepsis – this includes measuring blood lactate level, perform blood cultures, administration of intravenous antibiotics, fluids and oxygen and measuring hourly urine output.16
The Surviving Sepsis Campaign has also introduced Time zero or Time of Presentation (2018), which is defined as the time of triage in the emergency department or, if presenting from another care venue, from the earliest chart annotation consistent with all elements of sepsis to improve compliance and performance.16 The effect of morbidity and mortality in sepsis has been well documented and adopting sepsis care bundles has been shown to significantly reduce mortality.17
In sepsis-induced hypotension, 75% of patients survive with prompt recognition and management but with every hour’s delay this figure falls by >7%, implying that the mortality typically increases by approximately 30%.11 Longer-term studies show later death rates, one year after a bout of septic shock to exceed 50%18 and frequent functional impairment even in those who survive.19
Early identification of sepsis remains a significant challenge in critically ill patients. Research in biomarkers of sepsis such procalcitonin, C-reactive protein and cytokines (IL-6, IL-10) is being undertaken to aid early identification of sepsis and its association with various pathogens and foci of infection.
Blood cultures are routinely performed in most patients with sepsis and indeed, they form part of sepsis 6/Hour-1 bundle.16 However, false negative results are extremely common, especially if antibiotics have already been administered. There is also usually a delay of 48–72 hours before results of standard blood cultures and microbiological sensitivities become available to the treating clinician. Molecular assays that use polymerase chain reaction to target the ribosomal DNA sequences of common pathogens have been developed for clinical application, as the ability to detect bacterial DNA does not require the presence of live bacteria in the blood stream. While these techniques may allow rapid identification of bacterial pathogens they appear to lack adequate sensitivity to allow clinicians to reliably exclude bacterial infection as a cause of sepsis.20,21
Computed tomography (CT) with oral contrast remains the commonly used diagnostic modality to identify abdominal and pelvic sepsis, and is able to do so with high sensitivity and specificity in the hands of an experienced radiologist. However, safely moving a haemodynamically unstable patient out of the critical care environment into a CT scanner is never desirable and, on some occasions, might not be possible. The use of bedside diagnostic laparoscopy in such instances may reduce the risk involved in transferring unwell patients to a radiology department and also avoid the adverse consequences of negative laparotomy.22,23
Specifically, mortality after a negative bedside laparoscopy (18.3%) was also statistically lower than the percentage of patients who died after a negative exploratory laparotomy (38.9%) and the mean operating time in those patients who underwent laparotomy after bedside laparoscopy was lower, possibly because a ‘tailored laparotomy’ could then be undertaken.23
NICE guidelines emphasise rapid commencement of empirical antimicrobial therapy based on data showing a direct and strong relationship between the timeliness of appropriate therapy and survival.14 While early administration of appropriate antibiotics is, therefore paramount in the management of sepsis, audits of clinical practice in this area suggest there may be considerable room for improvement. The third patient report of the National Emergency Laparotomy Audit24 has examined patients with peritonitis and reported that <25% patients received their first dose of antibiotics within 1 hour and approximately 25% waited >6 hours from admission respectively.
The Surviving Sepsis Campaign8 reported that early antibiotic administration within three hours was independently associated with survival, but this was achieved in only 67% of cases.12 The recommendation has since been changed to delivery of antibiotics within one hour of diagnosis of sepsis12 but it seems even less likely that this will be achieved unless there is a sea change in clinical performance. NHS England have aimed to do this by incorporating targets for timely antibiotic treatment and review into the CQUIN (Commissioning for Quality and Innovation) payments framework for the care of patients in England, leading to powerful financial incentives for the delivery of optimum care.
Source identification and control
Early detection and timely therapeutic intervention can improve prognosis and overall clinical outcome in septic patients. Patients with severe sepsis are at greatest risk of developing septic shock. There is no direct evidence to confirm that delayed source control worsens outcome but it seems obvious that it will. There are obvious advantages to physically establishing control of a source of infection before progression to septic shock occurs, given the associated 5–10-fold rise in mortality which occurs as the patient deteriorates.25,26
Guidance on source control published in 201127 recommends that timing of intervention to control the source of sepsis, that is, surgery or equivalent (for example, radiological drainage) should be carried out in a time frame consistent with the severity of the clinical situation, so that, where relevant, source control should be undertaken immediately in patients with septic shock, within six hours in patients with organ dysfunction but without shock, and in less severely ill patients with uncomplicated sepsis within 18 hours. These targets were established prior to the most recent definitions of sepsis and it remains to be seen how they will be changed given more recent emphasis on qSOFA and NEWS2 scores as soon as possible.
Traditionally, drainage of collections has been achieved by surgical exploration. However, radiological drainage of abdominal abscesses using ultrasound was first described in 197428 and CT in 1977.29 While surgical intervention is still required where drainage of an abscess alone cannot achieve source control (for example in free visceral perforation or where resection or debridement are required), where radiological drainage cannot be undertaken adequately or safely (for example, where there are multiple and /or interloop abscesses), in many cases radiological drainage is possible and here it is the standard of care.29,30 Most authorities recommend draining all collections larger than 3cm in patients with systemic signs of infection.29,31
A national study conducted in the USA has reported that the number of percutaneous drainage procedures for intra-abdominal sepsis more than doubled between 2001 and 2013, while the laparotomy rate fell by 21% over the same period.30 Similar studies in patients with intra-abdominal abscesses due to Crohn’s disease noted that 29% could be managed by percutaneous drainage, whereas 32% still required surgery.32
Interventional techniques can also be used to treat other foci of sepsis. A study comparing clinical efficacy and adverse events of percutaneous cholecystostomy (PC) and early cholecystectomy (EC) in a large group of severely ill patients with acalculous cholecystitis (AC) has shown PC to be a safe and cost-effective bridging treatment strategy, with perioperative outcomes superior to those of early open cholecystectomy. Compared with open or laparoscopic EC, PC was superior in terms of morbidity, intensive-care unit admissions, length of hospital stay, and cost.33
Two studies have shown that PC is an effective procedure in seriously ill patients with AC and may be regarded as a definite treatment option in the majority of patients.34,35 Percutaneous drainage of liver and splenic abscesses can also be undertaken safely as first-line treatment prior to/instead of surgical intervention. A meta-analysis of five randomised, controlled trials comparing catheter drainage and repeated needle aspiration of liver abscesses has demonstrated catheter drainage to be more effective, with higher success and shorter time to achieve clinical improvement.36 Drainage of retroperitoneal collections can be performed transgastrically where required.
Drainage of pelvic collections can be challenging and in appropriate circumstances a transgluteal approach and endoscopically assisted ultrasound-guided drainage by transrectal, transperineal or transvaginal access is associated with a low risk of complications and should be considered for deep pelvic abscesses.37
European consensus guidelines recommend catheters of 7 – 10F for the treatment of most abscesses, regardless of abscess dimensions. However, large catheters (>10F) may be required for complex abscesses with thick contents.37 There are still no established guidelines for subsequent management of drains with regards to flushing the abscess, interval imaging and timing of removal. Regular, small-volume, gentle flushes should be used in such abscess cavities. Some centres advocate contrast-imaging prior to drain removal,38 while others recommend that drains be removed 48 hours after output has stopped, following repeat imaging.39
Surgery will be required where radiological source control is not possible and where a patient needs a definitive procedure, for example, resection or debridement of necrotic tissue, management of complete anastomotic dehiscence, proximal diversion of the gastrointestinal tract etc. It may also be safer in some circumstances to transfer an unstable patient to an operating theatre than to the radiology department to establish source control.
Timing and adequacy of surgical source control are vital. The third National Emergency Laparotomy Audit (NELA) reported on the outcome of approximately 4000 patients requiring laparotomy for peritonitis and found an average delay of eight hours to reach theatre after admission to hospital.24 If the diagnosis has been ascertained, a tailored approach should be undertaken with regards to the optimum approach and technique.
In dangerously unstable patients, a truncated ‘source/damage-control operation’ can be undertaken, in which abscesses are drained, bowel resected with the stapled ends left in situ and a planned re-laparotomy (RL) subsequently undertaken.40
Kim et al showed that 22% of patients undergoing emergency surgery for intra-abdominal catastrophes, of whom at least 65% had sepsis, required a further unplanned laparotomy. Risk factors for RL were identified to be peripheral arterial disease, alcohol abuse, body mass index >29, ischaemic bowel and an interval greater than 60 hours from the time of symptom onset to undertaking initial laparotomy. The presence of two or more risk factors was associated with a 55% RL rate, and three or more factors with an 83% RL rate. RL was associated with a fourfold increase in the rate of hospital death.41 In such situations, controversies arise regarding managing complex abdominal infection with an open abdomen technique that will allow re-look and also mitigate concerns of intra-abdominal hypertension.
While the only randomised controlled trial undertaken under these circumstances showed that the abdomen should be closed if it is safe to do so, attempts at primary closure in all circumstances may be associated with an increased incidence of multi-organ failure, resulting in poor survival. Conversely, the risk of treating abdominal sepsis with an open abdomen include significant disruption of respiratory mechanics, loss of abdominal ‘domain’, exposure to nosocomial pathogens, challenging wound care and even enteroatmospheric fistulation.Various temporary abdominal closure (TAC) techniques have been described, involving gauze and large, impermeable, self-adhesive membrane dressings; mesh (for example, Vicryl™, Dexon™); nonabsorbable mesh (for example, GORE-TEX™, polypropylene), negative pressure wound therapy (NPWT), NPWT with continuous fascial traction, dynamic retention sutures, Wittmann patch™, and Bogota bag.
A large systematic review of 74 studies in over 4300 patients (of which 79% had received treatment for peritonitis) showed that NPWT was the most frequent described TAC technique, and the highest weighted fascial closure rate was found in series describing NPWT with continuous mesh or suture mediated fascial traction and dynamic retention sutures.42 However, it was not possible to show differences in mortality, fistula and fascial closure rates, when comparing NPWT alone and NPWT with fascial traction.42
Laparoscopic treatment in abdominal sepsis is becoming more commonly used.22,23 However the most recent NELA report has not shown an increase in its use over a three-year period, which remained at 8%.24 Therapeutic advantages of laparoscopy are well known in the management of appendicitis, cholecystitis, and perforated gastric and duodenal ulcer.
In other situations such as diverticulitis, results from a multicentre randomised trial have not shown laparoscopic lavage to be superior to sigmoid resection for the treatment of purulent perforated diverticulitis.43 Great care should be taken when laparoscopic treatment is being considered for use in the septic abdomen, and should include assessment of the source of sepsis, the likelihood that adequate source control can be achieved by laparoscopic means, patient physiology and habitus, the risks of injury to other organs in a potentially hostile environment and, not least, the training and expertise of the surgical team. However, as technological advancement extends the potential range of minimally invasive procedures, it seems likely that wider adoption of these approaches will, at least in selected patients, reduce the need for open surgical treatment for abdominal sepsis.
1 Moynihan BG. The ritual of a surgical operation. Br J Surg 1920;8(29):27–35.
2 Bone RC et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992;101(6):1644–55.
3 Levy MM et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31(4):1250–6.
4 Shankar-Hari M et al. Developing a new definition and assessing new clinical criteria for septic shock: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016;315(8):775–87.
8 Rhodes A et al. The Surviving Sepsis Campaign bundles and outcome: results from the International Multicentre Prevalence Study on Sepsis (the IMPreSS study). Intensive Care Med 2015;41(9):1620–8.
9 Moore LJ et al. The epidemiology of sepsis in general surgery patients. J Trauma. 2011;70(3):672–80.
10 Moore LJ et al. Computerized clinical decision support improves mortality in intra abdominal surgical sepsis. Am J Surg 2010;200(6):839–43; discussion 43–4.
12 Soop M, Carlson GL. Recent developments in the surgical management of complex intra-abdominal infection. Br J Surg 2017;104(2):e65–e74.
13 Dellinger RP et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;41(2):580–637.
National Institute for Health and Care Excellence (NICE). Sepsis: recognition, diagnosis and early management: NICE Guidenline [NG51] 2016 [updated September 2017. www.nice.org.uk/guidance/ng51
(accessed July 2018).
17 Daniels R et al. The sepsis six and the severe sepsis resuscitation bundle: a prospective observational cohort study. Emerg Med J 2011;28(6):507–12.
18 Yende S, Angus DC. Long-term outcomes from sepsis. Curr Infect Dis Rep 2007;9(5):382–6.
19 Iwashyna TJ et al. Long-term cognitive impairment and functional disability among survivors of severe sepsis. JAMA 2010;304(16):1787–94.
20 Chang SS et al. Multiplex PCR system for rapid detection of pathogens in patients with presumed sepsis – a systemic review and meta-analysis. PLoS One 2013;8(5):e62323.
21 Warhurst G et al. Diagnostic accuracy of SeptiFast multi-pathogen real-time PCR in the setting of suspected healthcare-associated bloodstream infection. Intensive Care Med 2015;41(1):86–93.
22 Ceribelli C et al. Bedside diagnostic laparoscopy for critically ill patients: a retrospective study of 62 patients. Surg Endosc 2012;26(12):3612–5.
23 Alemanno G et al. Bedside diagnostic laparoscopy for critically ill patients in the Intensive Care Unit: Retrospective study and review of literature. J Minim Access Surg 2018; Feb 27 [Epub ahead of print].
NELA Project Team. Third Patient Report of the National Emergency Laparotomy Audit (NELA) RCoA London 2017. www.nela.org.uk/reports
(accessed July 2018).
25 Angus DC et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29(7):1303–10.
26 Annane D et al. Current epidemiology of septic shock: the CUB-Rea Network. Am J Respir Crit Care Med 2003;168(2):165–72.
28 Smith EH, Bartrum RJ. Ultrasonically guided percutaneous aspiration of abscesses. Am J Roentgenol Radium Ther Nucl Med 1974;122:308–12.
29 Rivera-Sanfeliz G. Percutaneous abdominal abscess drainage: a historical perspective. AJR Am J Roentgenol 2008;191(3):642–3.
30 Levin DC et al. Trends in use of percutaneous versus open surgical drainage of abdominal abscesses. J Am Coll Radiol 2015;12(12 Pt A):1247–50.
31 Robert B, Yzet T, Regimbeau JM. Radiologic drainage of post-operative collections and abscesses. J Visc Surg 2013;150(3 Suppl):S11–8.
32 Ananthakrishnan AN, McGinley EL. Treatment of intra-abdominal abscesses in Crohn’s disease: a nationwide analysis of patterns and outcomes of care. Dig Dis Sci 2013;58(7):2013–8.
33 Simorov A et al. Emergent cholecystostomy is superior to open cholecystectomy in extremely ill patients with acalculous cholecystitis: a large multicenter outcome study. Am J Surg. 2013;206:935–40; 40–1.
34 Chung YH, Choi ER, Kim KM. Can percutaneous cholecystostomy be a definitive management for acute acalculous cholecystitis? J Clin Gastroenterol 2012;46:216–9.
35 Kirkegard J, Horn T, Christensen SD. Percutaneous cholecystostomy is an effective definitive treatment option for acute acalculous cholecystitis. Scand J Surg 2015;104(4):238–43.
36 Cai Y, Xiong X, Lu J. Percutaneous needle aspiration versus catheter drainage in the management of liver abscess: a systematic review and meta-analysis. HPB (Oxford) 2015;17(3):195–201.
37 Dietrich CF et al. EFSUMB Guidelines on Interventional Ultrasound (INVUS), Part III – Abdominal Treatment Procedures (Long Version). Ultraschall Med 2016;37(1):E1–E32.
38 Feagins LA et al. Current strategies in the management of intra-abdominal abscesses in Crohn’s disease. Clin Gastroenterol Hepatol 2011;9(10):842–50.
39 Kassi F et al. Predictive factors for failure of percutaneous drainage of postoperative abscess after abdominal surgery. Am J Surg 2014;207(6):915–21.
40 Jansen JO, Loudon MA. Damage control surgery in a non-trauma setting. Br J Surg. 2007;94(7):789–90.
41 Kim JJ et al. Predictors of relaparotomy after nontrauma emergency general surgery with initial fascial closure. Am J Surg 2011;202(5):549–52.
42 Atema JJ, Gans SL, Boermeester MA. Systematic review and meta-analysis of the open abdomen and temporary abdominal closure techniques in non-trauma patients. World J Surg 2015;39(4):912–25.
43 Vennix S et al. Laparoscopic peritoneal lavage or sigmoidectomy for perforated diverticulitis with purulent peritonitis: a multicentre, parallel-group, randomised, open-label trial. Lancet. 2015;386(10000):1269–77.