Sepsis and septic shock: quick reference for the first hour

Purpose and scope

A quick reference for the doctor on call faced with suspected sepsis in adults: rapid recognition, microbiological sampling, empirical intravenous antibiotics, fluid, a vasopressor and early source control. Sepsis is a dynamic condition. Investigation and treatment must therefore proceed in parallel and must not wait for complete diagnostic certainty.

Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection. Clinically this usually corresponds to an acute increase in the SOFA score of at least 2 points. Septic shock is the subgroup that, despite the treatment of hypovolaemia, requires a vasopressor to maintain a MAP of at least 65 mmHg and at the same time has a lactate above 2 mmol/L. This combination carries a high mortality [1,2].

Give antibiotics immediately and aim for administration within one hour in septic shock, or when sepsis is judged certain or highly likely. With possible sepsis without shock, a rapid targeted assessment is made. If the suspicion persists, antibiotics must be given within three hours. The evidence for an absolute one-hour requirement is clearest in septic shock and less certain in sepsis without shock [3,4].

The first hour, in order

The measures should in practice overlap. Document the time of the suspicion of sepsis, the first dose of antibiotic, the cultures, the fluid bolus and the start of the vasopressor.

Sepsis alerts and standardised care processes can shorten the time to cultures, antibiotics, fluid and lactate measurement. A meta-analysis of emergency department alert systems found an association with lower mortality, but most of the evidence was observational and the results must not be taken to mean that an electronic alert replaces clinical judgement [5].

Recognising sepsis

Suspected or confirmed infection together with new organ dysfunction must raise the suspicion of sepsis. The patient need not have fever, leucocytosis or a positive culture. Older people, the immunosuppressed and patients with renal or hepatic failure can have subtle symptoms.

Findings that suggest acute organ dysfunction

An acute increase in the SOFA score of at least 2 points operationalises the organ dysfunction in Sepsis-3, but a full SOFA calculation must not delay treatment [1]. The qSOFA consists of a respiratory rate of at least 22/min, a systolic blood pressure of 100 mmHg or less, and an altered mental state. Two or three of these mark a high risk, but the qSOFA has inadequate sensitivity as a screening method on its own, and a low score does not exclude sepsis [3,6].

Important clinical warning signs

• New confusion can be the first sign in an older patient.
• Hypothermia, particularly in older or immunosuppressed patients, is a serious finding.
• Tachypnoea is often an early sign of metabolic acidosis or of pulmonary involvement.
• A normal blood pressure does not exclude hypoperfusion.
• A normal lactate does not exclude sepsis.
• A high lactate is not specific for sepsis. It can also be caused by adrenergic stress, seizures, hepatic impairment, hypoxaemia, poisoning and drugs.
• A patient who initially appears relatively well can deteriorate rapidly. Follow the vital signs and the clinical status serially.

Consider alternative diagnoses at the same time

Sepsis has no single diagnostic test. A substantial proportion of patients initially treated for sepsis turn out to have a non-infectious diagnosis. Important differential diagnoses are:

• haemorrhage or other hypovolaemia
• pulmonary embolism
• acute coronary syndrome or cardiogenic shock
• aortic dissection
• anaphylaxis
• adrenal insufficiency
• poisoning
• pancreatitis
• diabetic ketoacidosis
• thyrotoxic crisis
• a drug reaction
• bowel ischaemia
• inflammatory or haematological disease

Alternative diagnoses must be actively sought, but the investigation must not delay antibiotics and circulatory support in likely septic shock.

ABCDE and initial stabilisation

A, the airway

Assess whether the patient can protect the airway. Gurgling breathing, repeated aspiration, a rapidly falling level of consciousness or marked exhaustion require immediate contact with anaesthesia or intensive care.

Intubating a patient in septic shock carries a high risk of peri-intubation hypotension and cardiac arrest. Optimise the circulation, have noradrenaline running or immediately available, and plan the induction with an experienced anaesthetist.

B, breathing

Give oxygen if the patient is hypoxaemic. A reasonable initial target for most patients is a saturation of around 92 to 96 per cent. With known or likely chronic hypercapnic respiratory failure, a lower target is generally used, often 88 to 92 per cent, while the blood gas is followed. Avoid routine hyperoxia.

Assess:

• the respiratory rate and the work of breathing
• the saturation and the oxygen requirement
• an arterial or venous blood gas
• the auscultatory findings
• pulmonary oedema
• pneumonia, pleural effusion or pneumothorax
• the need for a high-flow nasal cannula, non-invasive ventilation or intubation

C, the circulation

Insert two peripheral cannulae, preferably large-bore and proximally placed. Attach continuous ECG monitoring and measure the blood pressure frequently. An arterial line is valuable during vasopressor treatment but must not delay the start of treatment.

Assess more than the blood pressure:

• the mental state
• capillary refill
• skin temperature and mottling
• the pulse and rhythm
• the urine output
• the lactate trend
• signs of fluid excess
• the response to a passive leg raise or a small fluid bolus

Capillary refill is a rapid and cost-free complement to lactate. In ANDROMEDA-SHOCK, a strategy guided by capillary refill did not give a statistically significant reduction in 28-day mortality compared with lactate-guided treatment, but it was associated with less organ dysfunction at 72 hours [7]. The lactate and the peripheral perfusion should therefore be interpreted together, not set against each other.

D, neurology and glucose

Document the GCS or RLS, the pupils and any focal neurological signs. Check the plasma glucose. Hypoglycaemia must be treated immediately. New impairment of consciousness can be due to sepsis-associated encephalopathy but at the same time requires assessment for, among other things, meningitis, encephalitis, stroke, hypercapnia, hypoxaemia and poisoning.

E, exposure and the focus

Examine the whole patient, including the back, the perineum, the skin folds, cannula sites and surgical wounds. Look particularly for:

• petechiae or purpura
• cellulitis or soft tissue infection
• pain out of proportion to the skin findings
• an infected vascular catheter
• septic arthritis
• pressure sores
• a surgical complication
• abdominal findings suggesting perforation, ischaemia or an abscess
• renal angle tenderness
• a gynaecological or obstetric focus

Investigations

Tests usually taken immediately

• a blood gas with lactate, pH, pCO2, bicarbonate and base excess
• a full blood count with a differential and platelet count
• CRP
• sodium, potassium, creatinine and urea
• liver function tests and bilirubin
• INR, APTT and fibrinogen in severe sepsis or where bleeding is suspected
• plasma glucose
• troponin and NT-proBNP where the differential diagnosis is relevant
• a urine dipstick and urine culture
• a pregnancy test where relevant
• targeted virological testing where it may affect isolation or treatment

Procalcitonin must not be used to decide whether to give initial antibiotics in clinically likely sepsis. The test has limited sensitivity and specificity and can give false reassurance early in the course. It can, however, contribute in some intensive care settings to the decision to stop antibiotics, together with clinical judgement. Meta-analyses show shorter antibiotic courses, but the evidence for a mortality benefit is uncertain [3,8].

Blood cultures

Take two complete sets of blood cultures before antibiotics. A set normally consists of one aerobic and one anaerobic bottle. Take the sets, where possible, from separate venepunctures. An adequate blood volume is decisive for the sensitivity.

Cultures must not delay antibiotics in septic shock. If the sampling cannot be completed quickly, the antibiotic is given anyway. International guidelines state that cultures before antibiotics are desirable provided this does not cause a substantial delay, of roughly more than 45 minutes [3].

Antibiotics clearly reduce the yield of cultures. In a systematic review, the pooled proportion of positive bacterial cultures fell from 37 per cent before antibiotics to 18 per cent after treatment had started [9]. Take cultures therefore only when this can be done without dangerous delay.

Take targeted cultures from the suspected focus:

• urine
• respiratory secretions
• a wound
• an abscess
• joint fluid
• pleural fluid
• cerebrospinal fluid
• the biliary tract or a drain
• the catheter tip after removal when catheter infection is suspected

Imaging

Imaging must answer a specific question and must not delay immediate treatment. Common investigations are:

• a chest radiograph or lung ultrasound
• ultrasound of the biliary tract and the urinary tract
• CT of the thorax and abdomen with intravenous contrast
• CT of the urinary tract where an obstructed infected kidney is suspected
• echocardiography in shock, with a murmur, with suspected endocarditis, or with cardiac dysfunction
• ultrasound of a joint or of soft tissue

Contrast-enhanced CT can be decisive in finding a drainable or surgical focus. The risk of delayed diagnosis and of failure to achieve source control often outweighs the risk of contrast-associated renal impairment in a patient in septic shock.

Empirical antibiotics

The time to antibiotics

In septic shock, antibiotics must be given immediately and preferably within one hour of the condition being recognised. A meta-analysis from 2025 covering 79,246 patients found lower mortality when antibiotics were given within one hour in septic shock. In sepsis without shock, no definite advantage was seen from an absolute one-hour limit, while treatment within three hours was associated with lower mortality [4].

Another meta-analysis of more than 106,000 patients found that each hour of delay was associated with higher mortality, but the heterogeneity was high and observational studies can be influenced by how time zero is defined and by sicker patients being managed differently [10]. The practical conclusion is:

Choice of agent

Base the choice on:

• the likely focus
• whether the infection is community-acquired or health care-associated
• previous culture results and resistance
• recent antibiotic treatment
• hospital admission, or health care received abroad
• the risk of ESBL producers, Pseudomonas or MRSA
• neutropenia or other immunosuppression
• drug allergy
• renal and hepatic function
• body weight
• the local resistance situation

The most important thing is that the first dose is adequate and given in time. A dose reduction for renal failure is often not appropriate for the first loading dose of a beta-lactam, since the volume of distribution is increased in severe sepsis. Subsequent dosing must, however, be adapted to renal function, the mode of dialysis, the weight and the clinical course.

Empirical treatment with an unclear focus

The table reproduces the doses given in the source material for this article from Strama Stockholm, empirical antibiotic treatment of adults in hospital 2025/2026. Always check the current local protocol, particularly since regional choices of agent, dosing intervals and recommendations on aminoglycosides may differ.

The doses below apply in general to an adult of normal weight with normal renal and hepatic function.

The agents listed for the same situation are alternatives, not drugs that should routinely be combined. Gentamicin is a possible initial addition according to Swedish local protocols in septic shock, but international guidelines do not recommend double Gram-negative cover routinely when the risk of multiresistant organisms is low [3]. Weigh the expected benefit against renal function, previous aminoglycoside exposure and other nephrotoxicity.

For unstable patients, according to the regional source cited, an extra dose of the beta-lactam is given halfway through the interval between the first and the second regular doses. After stabilisation the usual dosing interval is resumed. Follow the local protocol.

Gentamicin is dosed on the adjusted body weight in obesity:

Adjusted weight = ideal body weight + 0.4 × (actual weight minus ideal body weight).

At a creatinine clearance below 20 mL/min, another agent or specialist-determined dosing should be considered. A single dose carries less risk than continued aminoglycoside treatment, but it is not risk-free.

The beta-lactam as an extended infusion

After the first rapid loading dose, an extended or continuous infusion of the beta-lactam can be considered in intensive care, particularly in septic shock, with a high volume of distribution, with augmented renal clearance, or with organisms with high MIC values. A meta-analysis of 18 randomised trials with 9,108 intensive care patients found that extended infusion was associated with lower 90-day mortality and higher clinical cure than intermittent infusion [11]. Practical implementation depends on the stability of the agent and on local infusion procedures.

Penicillin allergy

Establish the type and timing of the reaction. Nausea, diarrhoea or a non-specific rash in childhood are not the same thing as immediate anaphylaxis. In septic shock, an erroneous label of allergy can lead to poorer empirical treatment.

With previous anaphylaxis, a severe mucocutaneous reaction or drug-induced organ injury, the choice of antibiotic must be discussed urgently with an infectious diseases physician and must follow the local protocol. Do not give an uncertain default alternative without taking the focus and the resistance risk into account.

Review the antibiotic

Carry out a structured antibiotic review within 24 to 48 hours at the latest, and daily thereafter:

1. Is there still a likely infectious focus?
2. Is the initial treatment active against the organism cultured?
3. Can the treatment be narrowed?
4. Does the dose need adjusting?
5. Is the source control adequate?
6. Can the antibiotic be stopped if infection is no longer likely?
7. How long a course does the specific focus require?

Antibiotics must not be continued merely because the patient once met the criteria of a sepsis alert. The cultures, the imaging, the clinical course and the alternative diagnoses must all be weighed together.

Fluid therapy

The initial strategy

Use a balanced crystalloid in the first instance, for example Ringer's acetate. International guidelines prefer a balanced crystalloid to sodium chloride 9 mg/mL, although the recommendation is weak [3,12]. Meta-analyses suggest a lower risk of mortality and acute kidney injury with balanced solutions, but the difference is less certain when only randomised trials are analysed [13].

Generally give 500 mL as a rapid bolus for hypotension or clinical hypoperfusion, and assess the effect immediately. In patients with heart failure, on dialysis, with marked hypoxia, right ventricular failure or pulmonary oedema, 250 mL is often more appropriate.

Assess after every bolus:

• the MAP and the systolic blood pressure
• capillary refill
• the mental state
• skin temperature and mottling
• the pulse
• the work of breathing and the saturation
• the chest findings on auscultation and ultrasound
• the urine output
• a change in stroke volume or pulse pressure
• a passive leg raise where this can be assessed reliably

Stop repeating boluses when the patient is no longer fluid-responsive or when signs of fluid overload appear.

How should 30 mL/kg be interpreted?

The benchmark of 30 mL/kg of crystalloid within three hours in sepsis-induced hypoperfusion or septic shock remains in international guidelines, but the recommendation is weak and rests largely on observational data [3,14]. It is not a requirement to give the whole volume without reassessment.

Fluid must be individualised in:

• heart failure
• severe renal failure or dialysis
• pulmonary oedema or marked hypoxaemia
• right ventricular failure
• hepatic cirrhosis
• a large volume already given prehospitally
• suspected bleeding
• another clear cause of hypovolaemia

The CLOVERS trial compared a restrictive strategy with earlier vasopressor use against a more liberal fluid strategy after the patients had already received 1 to 3 litres of fluid. Mortality did not differ, despite a difference of more than 2 litres in subsequent fluid administration [15]. The result supports giving priority to a vasopressor early when initial fluid does not correct the hypotension, but it does not say that an obviously hypovolaemic patient should be denied fluid.

Older Cochrane evidence lacked sufficient adult studies and at the same time showed that generous fluid can be harmful in some paediatric settings [16]. A fixed volume must therefore never replace repeated physiological assessment.

Avoid unsuitable fluids

• Do not use hydroxyethyl starch in sepsis.
• Gelatin is not recommended routinely.
• Albumin can be considered in intensive care when very large volumes of crystalloid have been required, but it is not the first-line fluid.
• Sodium chloride may be justified in particular situations, but large volumes increase the chloride load.
• With concomitant traumatic brain injury, different considerations apply from those in isolated sepsis [12].

Vasopressor and blood pressure targets

Noradrenaline is the first choice

Start noradrenaline if the MAP is below 65 mmHg despite initial fluid, if the patient is not judged able to tolerate more fluid, or if the hypotension is so marked from the outset that further waiting is dangerous. Fluid and a vasopressor can be given in parallel.

Noradrenaline has a better evidence base than dopamine and causes fewer arrhythmias. A meta-analysis found lower mortality and fewer serious adverse events with noradrenaline than with dopamine [17].

More recent pooled evidence suggests that early initiation of noradrenaline can achieve the MAP target faster, reduce fluid administration and possibly lower mortality. The evidence is not, however, conclusive, since there are few randomised trials and the results are influenced by observational data [18]. Do not therefore wait until several litres of fluid have been given if the patient is still clearly hypotensive.

Peripheral administration

Noradrenaline may be started in a well-functioning peripheral cannula. This is safer than allowing severe hypotension to persist while waiting for a central venous catheter.

Where possible use:

• a large-bore cannula in a large vein
• placement in the forearm or upper arm rather than the hand or foot
• a separate cannula for the vasopressor
• a visible puncture site
• frequent checks for pain, swelling, pallor and the function of the infusion
• a back-up cannula
• a local protocol for extravasation

A meta-analysis of more than 16,000 adults estimated the incidence of local complications at 1.8 per cent, but the heterogeneity was considerable and many studies were at risk of bias [19]. Another systematic review found extravasation in 3.4 per cent with no reported tissue necrosis or limb ischaemia [20]. Peripheral administration is therefore reasonable initially under careful monitoring. Central access is planned for a high dose, a prolonged requirement, poor peripheral access, or a need for several incompatible infusions.

The blood pressure target

The initial target is a MAP of at least 65 mmHg. At the same time, assess whether organ perfusion is improving. A higher numerical blood pressure is not a target in itself.

A higher target can be tried on an individual basis in:

• severe chronic hypertension
• persistent oliguria despite otherwise reasonable perfusion
• suspected inadequate cerebral perfusion
• particular neurological conditions

Randomised evidence shows no general mortality benefit from a higher MAP target. A higher target increases the risk of supraventricular arrhythmia, although patients with chronic hypertension may require renal replacement therapy less often [21]. Titrate therefore according to the clinical perfusion and the comorbidity, not routinely to 80 to 85 mmHg.

Additional vasopressors

With a rising noradrenaline requirement, vasopressin can be considered in intensive care according to the local protocol. Vasopressin often reduces the noradrenaline requirement but has not shown a definite mortality benefit. An individual patient data meta-analysis found no difference in 28-day mortality, fewer arrhythmias but more digital ischaemia [22].

Adrenaline can be used as an addition when the effect is inadequate. Where septic cardiomyopathy with a low cardiac output despite an adequate MAP is suspected, echocardiographic and intensive care assessment of inotropic treatment is required.

Hydrocortisone

Consider intravenous hydrocortisone in septic shock with a persisting vasopressor requirement despite reasonable fluid administration and noradrenaline. A common regimen in international guidelines is a total of 200 mg per day, either as 50 mg every six hours or as a continuous infusion [3]. Follow the local intensive care protocol for the starting criterion, the administration and the withdrawal.

An updated Cochrane review with 87 studies and more than 24,000 participants found that corticosteroids probably reduce 28-day mortality and hospital mortality slightly and may shorten the intensive care stay. The effect on long-term mortality was uncertain [23]. Hydrocortisone does not replace antibiotics, source control, fluid or a vasopressor.

Source control

Antibiotics cannot compensate for an undrained or devitalised focus. Identify early whether the patient needs a procedure.

Examples:

• an abscess
• an empyema
• an obstructed infected kidney
• cholangitis with obstruction
• a perforated hollow viscus
• an anastomotic leak
• a necrotising soft tissue infection
• septic arthritis
• an infected vascular catheter
• an infected prosthesis
• endocarditis with a complication
• ischaemic or necrotic tissue

Contact the relevant operating specialty, the interventional radiologist or the endoscopist immediately. International guidelines recommend that necessary source control be achieved as soon as it is medically and logistically possible, often with a practical target of within 6 to 12 hours [3]. The evidence for an exact time limit is limited and the studies use varying definitions, but the available data consistently indicate better outcomes when adequate source control is genuinely achieved [24].

Where catheter infection is suspected, the need for immediate removal of the catheter must be assessed. Establish new secure access first if the catheter is needed for a life-sustaining infusion.

Monitoring during the first few hours

Clinical targets

Follow at least:

• the MAP and the vasopressor dose
• the pulse and rhythm
• the saturation and the oxygen requirement
• the respiratory rate and the work of breathing
• the level of consciousness
• capillary refill
• skin temperature and mottling
• the hourly urine output
• the fluid balance
• the lactate trend
• the pH and any metabolic acidosis
• the creatinine, potassium and glucose

Insert a urinary catheter in shock or when accurate hourly urine measurement is needed. A guide value is a urine output of at least 0.5 mL/kg/h, but single hourly values must be interpreted with care. Oliguria can be due to hypoperfusion, acute kidney injury, obstruction or chronic renal failure, and must not automatically be treated with repeated fluid boluses.

Repeat the lactate

Repeat the lactate when the initial value was raised or the clinical picture deteriorates. The timing is individualised, often within 2 to 4 hours, and earlier with a rapid course. A falling lactate is favourable but must not on its own drive further fluid administration. A persisting elevation can be due to:

• continuing hypoperfusion
• adrenergic stimulation
• hepatic impairment
• seizures
• regional ischaemia
• drugs or toxins

Do not give fluid merely to normalise a lactate if the clinical examination indicates good perfusion or fluid excess.

Contact with intensive care

Contact intensive care early with:

• a need for noradrenaline
• a persisting MAP below 65 mmHg
• a rapidly rising lactate
• marked metabolic acidosis
• an increasing oxygen requirement
• a need for a high-flow nasal cannula, non-invasive ventilation or intubation
• a falling GCS
• oliguria with circulatory instability
• multiple organ dysfunction
• a need for urgent source control
• uncertainty about further fluid or the haemodynamics

Transfer must not delay antibiotics or the initial vasopressor. Delayed access to intensive care resources is associated with worse outcomes in observational studies [3].

Red flags and pitfalls

Foci requiring source control

An abscess, an empyema, an obstructed infected kidney, cholangitis with obstruction, a necrotising soft tissue infection, an infected catheter or an infected prosthesis all require more than antibiotics. Contact the surgeon, the urologist, the interventional radiologist or another relevant specialty urgently.

Necrotising soft tissue infection

Consider the diagnosis with:

• pain out of proportion to the skin findings
• rapid progression
• oedema extending beyond the erythema
• skin discoloration or bullae
• loss of sensation
• crepitus
• shock without a clear focus

A normal or moderately raised CRP early on does not exclude the condition. Do not wait for the classical skin findings to develop. Surgical exploration is decisive.

Meningitis or encephalitis

With headache, neck stiffness, petechiae, focal neurological signs, seizures or impaired consciousness, meningitis or encephalitis must be managed as a separate emergency. Give dexamethasone and antibiotics according to the local meningitis protocol without waiting for CT or lumbar puncture if these investigations would delay treatment. Add antiviral treatment when herpes encephalitis is a relevant possibility.

Neutropenia

Recent chemotherapy, haematological malignancy or known neutropenia change the choice of antibiotic. Fever may be absent and the local signs of inflammation may be subtle. Follow the separate neutropenic fever protocol.

Immunosuppression

Transplantation, high-dose steroids, biological agents, advanced HIV infection and other immune defects broaden the spectrum of possible pathogens. Contact an infectious diseases physician early, but do not delay initial broad treatment in shock.

Pregnancy and the postpartum period

Consider pyelonephritis, chorioamnionitis, endometritis, septic abortion and invasive streptococcal infection. Involve the obstetrician early. Normal pregnancy physiology affects the pulse, the blood pressure and the laboratory values.

Endocarditis

Consider endocarditis with:

• a new murmur
• embolic phenomena
• intravenous drug use
• a valve prosthesis or intracardiac device
• prolonged fever
• S. aureus in a blood culture
• unexplained bacteraemia

Take several sets of blood cultures if this can be done without delaying treatment, and plan echocardiography.

A normal lactate does not exclude sepsis

The lactate is a measure of risk and perfusion, not a diagnostic answer. Severe sepsis can be present with a normal lactate, particularly with isolated respiratory, renal, neurological or haematological organ dysfunction.

A high lactate does not always mean more fluid

Fluid is given for probable fluid-responsive hypoperfusion, not for a laboratory value in isolation. Repeated boluses without improvement increase the risk of pulmonary oedema, tissue oedema and renal venous congestion.

Do not wait for a central venous catheter

A peripheral noradrenaline infusion is acceptable initially with suitable access and monitoring [19,20]. An untreated MAP of 45 mmHg is more dangerous than a brief, controlled peripheral infusion.

Cultures must not delay antibiotics in shock

Take cultures only when this can be done quickly. Give the antibiotic immediately if the sampling is taking time. Document why the cultures were taken after treatment had started.

Do not forget the correct first dose

Underdosing is a risk in sepsis because of the large volume of distribution and the altered pharmacokinetics. Give the full loading dose unless there is a specific contraindication. Adapt the dose and interval thereafter to renal function and the clinical course.

Avoid unnecessary antibiotics in uncertain sepsis without shock

A one-hour requirement for everyone with a possible infection and abnormal vital signs risks overtreatment. In a stable patient without shock there is room for up to three hours of rapid diagnostic assessment if the infection is uncertain [3,4]. The monitoring must be active and antibiotics must be given at once if the likelihood or the severity increases.

Reconsider the diagnosis

If the patient does not improve, ask:

• Is the antibiotic the right one?
• Is the dose adequate?
• Is there an undrained focus?
• Is the diagnosis in fact haemorrhage, pulmonary embolism, cardiogenic shock, anaphylaxis or poisoning?
• Is there obstructive shock or bowel ischaemia?
• Has the patient developed a pneumothorax, tamponade or another complication?

A sepsis bundle does not replace clinical judgement

Systematic care bundles are associated with better process measures and survival, but there is no high-quality evidence that exactly the same antibiotic timing, fluid volume and lactate schedule suit every patient [25]. Standardise what must happen quickly, but individualise the volume, the blood pressure target and the continuing treatment.

A practical checklist before handover

• The time of the suspicion of sepsis documented
• ABCDE performed and repeated
• Two working cannulae
• A blood gas and lactate taken
• Two sets of blood cultures taken, or the treatment-related deviation documented
• Cultures from the focus taken where relevant
• Empirical antibiotic given, with the agent and time documented
• Allergy and previous resistance findings checked
• Fluid boluses given with the response documented
• Noradrenaline started for persisting or marked hypotension
• A MAP target prescribed
• Urine measurement established in shock
• Source control assessed and the relevant specialty contacted
• Intensive care contacted in shock or where advanced organ support is needed
• A plan for a repeat lactate and clinical reassessment documented
• A plan for an antibiotic review within 24 to 48 hours documented

References

1. Singer et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016. PMID: 26903338
2. Shankar-Hari 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. PMID: 26903336
3. Evans et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Medicine 2021. PMID: 34599691
4. Ku et al. Appropriate timing of antibiotic initiation in patients with sepsis or septic shock: a systematic review and meta-analysis. Korean Journal of Internal Medicine 2025. PMID: 40859809
5. Kim et al. Sepsis Alert Systems, Mortality, and Adherence in Emergency Departments: A Systematic Review and Meta-Analysis. JAMA Network Open 2024. PMID: 39037814
6. Seymour et al. Assessment of Clinical Criteria for Sepsis: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016. PMID: 26903335
7. Hernández et al. Effect of a Resuscitation Strategy Targeting Peripheral Perfusion Status vs Serum Lactate Levels on 28-Day Mortality Among Patients With Septic Shock: The ANDROMEDA-SHOCK Randomized Clinical Trial. JAMA 2019. PMID: 30772908
8. Pepper et al. Procalcitonin-Guided Antibiotic Discontinuation and Mortality in Critically Ill Adults: A Systematic Review and Meta-analysis. Chest 2019. PMID: 30772386
9. Itaeli et al. Rate of bacterial positivity following antibiotic initiation: a systematic review and meta-analysis. Systematic Reviews 2025. PMID: 41422045
10. Huang et al. The association between mortality and door-to-antibiotic time: a systematic review and meta-analysis. Postgraduate Medical Journal 2023. PMID: 36917816
11. Abdul-Aziz et al. Prolonged vs Intermittent Infusions of β-Lactam Antibiotics in Adults With Sepsis or Septic Shock: A Systematic Review and Meta-Analysis. JAMA 2024. PMID: 38864162
12. Tseng et al. Resuscitation fluid types in sepsis, surgical, and trauma patients: a systematic review and sequential network meta-analyses. Critical Care 2020. PMID: 33317590
13. Beran et al. Balanced Crystalloids versus Normal Saline in Adults with Sepsis: A Comprehensive Systematic Review and Meta-Analysis. Journal of Clinical Medicine 2022. PMID: 35407578
14. Prescott and Ostermann. What is new and different in the 2021 Surviving Sepsis Campaign guidelines. Medizinische Klinik, Intensivmedizin und Notfallmedizin 2023. PMID: 37286842
15. Shapiro et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. New England Journal of Medicine 2023. PMID: 36688507
16. Li et al. Liberal versus conservative fluid therapy in adults and children with sepsis or septic shock. Cochrane Database of Systematic Reviews 2018. PMID: 30536956
17. Avni et al. Vasopressors for the Treatment of Septic Shock: Systematic Review and Meta-Analysis. PLoS One 2015. PMID: 26237037
18. Shi et al. Early norepinephrine for patients with septic shock: an updated systematic review and meta-analysis with trial sequential analysis. Critical Care 2025. PMID: 40329359
19. Owen et al. Adverse events associated with administration of vasopressor medications through a peripheral intravenous catheter: a systematic review and meta-analysis. Critical Care 2021. PMID: 33863361
20. Tian et al. Safety of peripheral administration of vasopressor medications: A systematic review. Emergency Medicine Australasia 2020. PMID: 31698544
21. Yoshimoto et al. Optimal target blood pressure in critically ill adult patients with vasodilatory shock: A systematic review and meta-analysis. Frontiers in Physiology 2022. PMID: 36051909
22. Nagendran et al. Vasopressin in septic shock: an individual patient data meta-analysis of randomised controlled trials. Intensive Care Medicine 2019. PMID: 31062052
23. Annane et al. Corticosteroids for treating sepsis in children and adults. Cochrane Database of Systematic Reviews 2025. PMID: 40470636
24. Tabah et al. Source control in bloodstream infections in patients with sepsis, septic shock, or requiring ICU admission: a scoping review with recommendations for standardizing research. Intensive Care Medicine 2025. PMID: 40658248
25. Pepper et al. Antibiotic- and Fluid-Focused Bundles Potentially Improve Sepsis Management, but High-Quality Evidence Is Lacking for the Specificity Required in the Centers for Medicare and Medicaid Service's Sepsis Bundle (SEP-1). Critical Care Medicine 2019. PMID: 31369426