Patients with severe sepsis should be managed in an ICU or HDU setting. Good basic medical care is of paramount importance, as is close attention to infection control and prevention of hospital acquired infection.
Appropriate venous thromboembolism and stress ulcer prophylaxis should be given. If mechanically ventilated limitation of tidal volumes to 6ml/kg and airway plateau pressures to 30 cmH20 is recommended. Early enteral nutrition should be commenced once haemodynamic stability is achieved, though caution in those on very high doses of inotropes.
There are several additional therapies which have been advocated in severe sepsis which may improve outcome. Some of these have proved to be controversial.
Recombinant Human Activated Protein C (rh-APC)
rh-APC was found not to be beneficial to patients with severe sepsis/septic shock in the Prowess-Shock study. It's use is no longer recommended and it has been withdrawn.
Low Dose Steroids
Low dose steroids have been in use in severe sepsis for over 20 years, controversy surrounds their use. In the last 10 years the concept of relative adrenal insufficiency was postulated i.e. a subgroup of sepsis patients with an inadequate adrenal response, diagnosed with a corticoptrophin test. The largest trial to date is CORTICUS, a study of 500 patients, which did not demonstrate a survival benefit but did demonstrate more rapid shock reversal. Use of steroids should be balanced against potential complications which include infection, hyperglycaemia and weakness.
Insulin has anti-inflammatory properties and prevention of hyperglycaemia has been associated with a reduction in duration of ICU admission, acute renal failure and critical illness polyneuropathy. The optimum range for blood glucose is yet to be fully elucidated; two studies from Leuven in Belgium had suggested that tight glucose control (4.5-6.1 mmol/l) was associated with the best outcome, however a large multicentre RCT from Australia (NICE-SUGAR) has suggested that tight control was associated with an increased mortality, an unacceptable incidence of hypoglycaemic events and that less stringent control (e.g. < 10 mmol/l) has the same benefits.
Renal Replacement Therapy
Renal replacement therapy should be offered to patients with acute kidney injury if appropriate. There is currently no evidence favouring intermittent haemodialysis over continuous haemofiltration. The use of “high dose” haemofiltration had once been hypothesised to remove pro-inflammatory mediators with potential benefit; however there is currently insufficient evidence to support the use of this therapy. Two recent randomised controlled trials have failed to show benefit using “higher dose” RRT.