What late-onset sepsis is
Late-onset neonatal sepsis is defined as sepsis presenting more than 72 hours after birth. It is distinguished from early-onset sepsis, which arises within the first 72 hours and typically reflects vertical transmission from the maternal genital tract. Late- onset sepsis may arise from maternal colonisation, from the hospital environment, or from the infant’s own emerging commensal flora, and its causative organisms reflect this broader range of sources.
In preterm neonates — particularly those born before 32 weeks or weighing less than 1,500 grams — late-onset sepsis is one of the most significant causes of morbidity and mortality. The immature immune system, compromised skin and mucosal barriers, frequent invasive procedures, and prolonged hospital stay all increase susceptibility. Late-onset sepsis complicates between 15 and 30 percent of very-low-birthweight infant admissions in published neonatal unit audit series.
GBS prevalence and presentation
Group B Streptococcus (Streptococcus agalactiae, or GBS) is colonised in 10 to 30 percent of pregnant women in the United Kingdom, making it one of the most common organisms in the maternal peripartum environment. Late-onset GBS sepsis, presenting after 72 hours of life, is caused by horizontal acquisition from the environment or by delayed expression of vertically transmitted colonisation. It differs from early- onset GBS disease in that it is not reliably prevented by intrapartum antibiotic prophylaxis.
The clinical presentation of late-onset GBS sepsis is characterised by sudden onset of apnoea, bradycardia and circulatory collapse in a previously stable or relatively stable preterm infant. Temperature instability, poor perfusion, and metabolic acidosis develop rapidly. The speed and severity of deterioration can be extreme: infants who appeared clinically adequate on one assessment may be in profound septic shock within one to two hours. This clinical picture is indistinguishable at presentation from the deterioration trajectory that the Crown alleged was the result of deliberate harm in multiple indicted cases.
Enterovirus and parechovirus, which circulate in healthcare environments and can produce severe neonatal sepsis-like presentations, represent an additional infectious differential that has been raised in the context of the unit environment at the Countess of Chester Hospital during the relevant period.
Why it routinely escapes confirmation
The methodological difficulty with late-onset sepsis as a differential diagnosis is that microbiological confirmation requires a positive blood culture, and blood cultures have significant false-negative rates in neonatal sepsis. Blood culture sensitivity in neonatal sepsis is estimated at 70 to 80 percent under optimal conditions, and substantially lower when samples are taken after antibiotic administration has already begun. In a resuscitation scenario, blood cultures may be taken at a suboptimal time, in inadequate volume, or after empirical antibiotics have already been given.
The consequence is that a substantial minority of neonatal sepsis episodes — those meeting clinical criteria but with negative cultures — are classified as “culture-negative sepsis” or attributed to other causes. The absence of a positive blood culture does not exclude late-onset GBS sepsis and should not be treated as ruling out an infectious cause. This is standard teaching in neonatal medicine and is reflected in the NICE guideline on neonatal infection.
Dr Peter Donnelly’s microbiological commentary on the Countess of Chester case has addressed the infection-control and unit-environment picture, including the conditions under which late-onset sepsis could account for multiple deteriorations on the same unit within a defined time period.
What the indicted-case records show
Several of the infants whose cases formed the basis of criminal charges experienced acute deteriorations in the clinical setting of possible or probable late-onset sepsis. The contemporaneous records in these cases document empirical antibiotic commencement, clinical features consistent with sepsis, and in some instances inflammatory markers elevated in the period around the deterioration. The Panel’s case-by-case review has examined these records and identified infectious aetiology as a live and in some cases probable natural-cause explanation for the deteriorations.
Supporters of a CCRC referral note that the prosecution’s expert evidence did not systematically address the sepsis differential in each count. Dr Dewi Evans’s evidence focused on the alleged mechanism of deliberate harm without providing the jury with a structured evaluation of why infectious aetiology was excluded in each case. Where blood cultures were negative, the prosecution treated this as confirmation of a non-infectious cause, a reasoning error that does not accord with the recognised sensitivity limitations of blood culture in neonatal sepsis.
What independent paediatric review has concluded
Independent paediatric review, including the Panel led by Dr Shoo Lee, has found that late-onset sepsis cannot be excluded as the primary cause of deterioration in multiple indicted cases. The Panel’s conclusions identify infectious aetiology as consistent with the clinical and microbiological picture in those cases, and note that the prosecution’s approach to the sepsis differential was inadequate to support the exclusion that its argument required.
The broader infection-control context is also relevant. The Countess of Chester neonatal unit experienced a period of elevated mortality over the relevant years. A cluster of deteriorations on a single unit in a defined time window raises, as a primary hypothesis, the possibility of a unit-level infectious or environmental factor. Independent microbiological review has examined this hypothesis and found it consistent with the observable pattern, independently of any question about Letby’s role.