Background Cerebral abscess is a rare but recognised complication in cyanotic congenital heart disease (CHD) with a poor prognosis, especially if diagnosis is delayed1. Method: Retrospective analysis of paper and electronic records of ACHD patients at Leeds General Infirmary (LGI) with a history of confirmed cerebral abscess.

Results 7 cases of cerebral abscess were managed in LGI 1981–2019 (table 1). One occurred prior to transition to adult services, mean age 34 years. All had cyanotic CHD, and all but 1 occurred prior to any cardiac intervention. All presented with either focal neurology or headache, with other symptoms in graph 1. Only 1 had normal inflammatory markers on admission and there was no evidence of vegetations on any echocardiograms. One patient had suture diastasis on skull x-ray, and all had a CT scan confirming ring enhancing lesions. One had multiple lesions and the location of the lesions was varied (parietal, frontal, thalamic and occipital). Three (43%) had streptococcus milleri on abscess aspirate with another documented as gram positive cocci. One had strep intermedius on blood cultures (member of streptococcus milleri group), but all other blood cultures demonstrated no growth. All underwent aspiration or surgical drainage of the abscess, 3 occurred on the day of diagnosis, and the longest duration prior to intervention was 13 days. All received a prolonged course of antibiotics of varying combinations (graph 2). Two developed seizures post-operatively and one developed a renal abscess. One patient had long-term neurological sequelae (Epilepsy and left homonymous hemianopia), one died 8 days post-operatively and another died 3 years later from end stage heart failure. Discussion: Streptococcus milleri group are recognised commensals of the oropharynx and gastrointestinal tract, and the predominant microbiological diagnosis in cerebral abscesses1. Yet only one case had recent dental work, and none had confirmed endocarditis by Duke’s criteria. Despite documented high rates of peri-operative complications2, in our cohort there was one death and one long-term neurological sequelae. It is imperative the diagnosis is confirmed at the earliest opportunity, with prompt neurosurgical referral regardless of underlying cardiac pathology and degree of cyanosis as these high risk patients have demonstrated good long-term post-operative outcomes.We acknowledge the limitations that paediatric cases were not reviewed, and that clinical practice has progressed since the earliest case in 1981.

Conclusion Prognosis has improved in recent decades, but CHD remains a recognised risk factor for death3. It is an important differential to exclude in any patient with cyanotic CHD presenting with a headache or focal neurology by performing urgent cerebral imaging4. Clinical management relies on retrospective studies and previous clinical experience as no trials report clinically meaningful outcomes for antibiotic use in cyanotic congenital heart disease with cerebral abscess owing to the rarity of the condition5.

Conflict of Interest None