Discussion

In this cohort study spanning over two decades, we found that premorbid HTN was associated with increased severity of the initial bleeding event as assessed by all measures of haemorrhage on the initial CT. Premorbid HTN was also associated with twice the frequency of aneurysm rebleeding, a risk that has not been widely recognised.

The association of premorbid HTN with more severe initial bleeding differs from a smaller study that did not confirm this association.11 Improved quality of urgent CT scans and the use of standardised scales to estimate the amount of blood in different intracranial compartments may explain this discrepancy. Higher blood pressure at ictus, or more advanced vessel structural damage due to atherosclerotic disease, might also explain the more severe burden of haemorrhage among SAH patients with premorbid HTN.

The association between premorbid HTN and rebleed retained its significance even after statistical adjustment for admission Hunt–Hess grade and size of the ruptured aneurysm—two previously established predictors of rebleeding in our database2—as well as age and aneurysm rupture site. Time from admission to aneurysm securing was equally short (1 day) in patients with and without premorbid HTN, and is therefore unlikely to explain the observed difference in aneurysm rebleeding. Rebleeding occurred in 8.6% of our study cohort, and should be viewed as a distinct and preventable cause of poor outcome in SAH patients with premorbid HTN. Our findings have implications for clinical practice. First, in all patients—particularly in those with premorbid HTN —the treating staff should closely monitor and control HTN as early as possible and, at least, as long as the ruptured aneurysm is unsecured. Second, a randomised controlled trial and single-centre cohort study indicate that early antifibrinolytic therapy until definitive aneurysm repair, up to 72 h after onset, reduces the risk of rebleeding by approximately 80%.12 ,13 Premorbid HTN and poor clinical grade imply a higher than average rebleeding risk, and should encourage clinicians to strongly consider this intervention whenever definitive aneurysm repair cannot be performed immediately.

Premorbid HTN was associated with an increased frequency of a variety of neurological and medical complications which most likely reflected worse clinical grade and older age in these patients. Surprisingly, premorbid HTN was not associated with an increased risk of DCI or angiographic vasospasm by contrast with other reports claiming a link.14 It is difficult to explain why the larger amounts of cisternal blood found in hypertensive patients were not associated with DCI. One putative mechanism is that patients with history of HTN had higher in-hospital blood pressure levels that may have reduced the risk of DCI. This remains a speculative explanation because the blood pressure levels during the hospitalisation were not available for this analysis. Alternatively, the association between premorbid HTN and cisternal blood was not sufficiently strong to translate into a higher rate of DCI. Of interest, aneurysm rebleeding was not associated with DCI in a previous publication from the same team.3 At 3 months, premorbid HTN was associated with a doubling of the mortality rate. This association remained significant after controlling for age, Hunt–Hess grade on admission and aneurysm size, the three most important baseline determinants of outcome in our database. This is in line with other studies that have linked premorbid HTN to poor outcome after SAH.11 ,15 This study has limitations. First, sedative medication or intubation can influence clinical grading, and data in this regard were lacking. Yet, sedation and intubation of SAH patients followed the same prespecified algorithm,10 so that we deem unlikely that a difference in treatment between patients with, versus without premorbid HTN, biased the reported clinical grading. Second, premorbid and in-hospital blood pressure levels after admission were not systematically available, so that we cannot exclude that differences in the actual blood pressure may have contributed to the different rebleeding rates. The higher in-hospital use of antihypertensive medication in patients with premorbid HTN suggests indeed that their in-hospital blood pressure levels were higher. The conclusions of the study, therefore, apply only to a ‘history of premorbid HTN’ as elicited in the emergency department. Third, the radiological scans were rated by neurocritical care faculty members and fellows who had full access to all medical data, so that blinding to the presence of premorbid HTN cannot be guaranteed. However, the rating of the radiological scans based on the prespecified, referenced criteria and information on premorbid HTN was not routinely incorporated into imaging rating. Finally, complete medical histories were lacking in 12% of patients in this observational study (only patients with complete medical histories were included).

In conclusion, a history of premorbid HTN represents a significant risk factor for aneurysm rebleeding. Given that aneurysm rebleeding is a potentially fatal—but preventable—complication, these findings are of clinical relevance.